Skip to content

perf(sort): continues sort and merge every x batches in background - #26527

Closed
rluvaton wants to merge 2 commits into
apache:mainfrom
rluvaton:sort-background
Closed

perf(sort): continues sort and merge every x batches in background#26527
rluvaton wants to merge 2 commits into
apache:mainfrom
rluvaton:sort-background

Conversation

@rluvaton

@rluvaton rluvaton commented Oct 20, 2026

Copy link
Copy Markdown
Member

Which issue does this PR close?

  • Closes #.

Rationale for this change

What changes are included in this PR?

What is the testing strategy for this PR?

Are there any user-facing changes?

@rluvaton

rluvaton commented Oct 20, 2026

Copy link
Copy Markdown
Member Author

run benchmark sort
env:
BENCH_FILTER: sort_axis/sort.*.1M.High.cardinality.Unsorted..100cols

@github-actions github-actions Bot added documentation Improvements or additions to documentation sqllogictest SQL Logic Tests (.slt) common Related to common crate execution Related to the execution crate physical-plan Changes to the physical-plan crate labels Oct 20, 2026
@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6290368015-3632-tjdjm 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort
env:
  BENCH_FILTER: "sort_axis/sort.*.1M.High.cardinality.Unsorted..100cols"

Results will be posted here when complete

File an issue against this benchmark runner

@rluvaton

rluvaton commented Oct 20, 2026

Copy link
Copy Markdown
Member Author

run benchmarks sort_tpch sort_tpch tpch tpch10 tpcds

env:
   prefer_hash_join: false

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6290446562-3634-xwktn 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpch
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6290446562-3635-77l6g 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpch10
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6290446562-3633-skt55 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort_tpch
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6290446562-3636-w2cr7 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpcds
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpch
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and sort-background
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ sort-background ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 40.37 ms │        40.20 ms │ no change │
│ QQuery 2  │ 18.12 ms │        17.96 ms │ no change │
│ QQuery 3  │ 28.00 ms │        28.27 ms │ no change │
│ QQuery 4  │ 17.01 ms │        17.08 ms │ no change │
│ QQuery 5  │ 34.86 ms │        35.01 ms │ no change │
│ QQuery 6  │ 13.51 ms │        13.33 ms │ no change │
│ QQuery 7  │ 41.17 ms │        41.07 ms │ no change │
│ QQuery 8  │ 39.97 ms │        39.85 ms │ no change │
│ QQuery 9  │ 48.72 ms │        49.70 ms │ no change │
│ QQuery 10 │ 39.81 ms │        39.92 ms │ no change │
│ QQuery 11 │ 13.31 ms │        12.83 ms │ no change │
│ QQuery 12 │ 17.25 ms │        17.68 ms │ no change │
│ QQuery 13 │ 38.71 ms │        38.61 ms │ no change │
│ QQuery 14 │ 24.47 ms │        24.34 ms │ no change │
│ QQuery 15 │ 31.13 ms │        30.46 ms │ no change │
│ QQuery 16 │ 13.52 ms │        13.40 ms │ no change │
│ QQuery 17 │ 73.28 ms │        73.17 ms │ no change │
│ QQuery 18 │ 44.71 ms │        45.31 ms │ no change │
│ QQuery 19 │ 31.59 ms │        31.74 ms │ no change │
│ QQuery 20 │ 31.45 ms │        31.86 ms │ no change │
│ QQuery 21 │ 55.46 ms │        54.91 ms │ no change │
│ QQuery 22 │ 13.82 ms │        13.34 ms │ no change │
└───────────┴──────────┴─────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary              ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)              │ 710.23ms │
│ Total Time (sort-background)   │ 710.05ms │
│ Average Time (HEAD)            │  32.28ms │
│ Average Time (sort-background) │  32.28ms │
│ Queries Faster                 │        0 │
│ Queries Slower                 │        0 │
│ Queries with No Change         │       22 │
│ Queries with Failure           │        0 │
└────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and sort-background
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃                sort-background ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 40.37 / 41.01 ±0.74 / 42.40 ms │ 40.20 / 40.91 ±1.01 / 42.91 ms │     no change │
│ QQuery 2  │ 18.12 / 18.31 ±0.31 / 18.92 ms │ 17.96 / 18.19 ±0.32 / 18.83 ms │     no change │
│ QQuery 3  │ 28.00 / 28.95 ±1.17 / 31.11 ms │ 28.27 / 28.51 ±0.17 / 28.73 ms │     no change │
│ QQuery 4  │ 17.01 / 17.22 ±0.19 / 17.51 ms │ 17.08 / 17.47 ±0.49 / 18.44 ms │     no change │
│ QQuery 5  │ 34.86 / 35.40 ±0.44 / 36.07 ms │ 35.01 / 35.04 ±0.04 / 35.11 ms │     no change │
│ QQuery 6  │ 13.51 / 13.62 ±0.11 / 13.83 ms │ 13.33 / 13.55 ±0.15 / 13.81 ms │     no change │
│ QQuery 7  │ 41.17 / 41.78 ±0.75 / 43.19 ms │ 41.07 / 42.25 ±1.00 / 43.79 ms │     no change │
│ QQuery 8  │ 39.97 / 40.70 ±0.70 / 41.97 ms │ 39.85 / 40.37 ±0.33 / 40.79 ms │     no change │
│ QQuery 9  │ 48.72 / 50.32 ±1.02 / 51.43 ms │ 49.70 / 50.37 ±0.64 / 51.59 ms │     no change │
│ QQuery 10 │ 39.81 / 39.98 ±0.15 / 40.24 ms │ 39.92 / 40.70 ±0.97 / 42.60 ms │     no change │
│ QQuery 11 │ 13.31 / 13.43 ±0.13 / 13.65 ms │ 12.83 / 13.25 ±0.29 / 13.71 ms │     no change │
│ QQuery 12 │ 17.25 / 17.74 ±0.30 / 18.14 ms │ 17.68 / 18.14 ±0.53 / 19.13 ms │     no change │
│ QQuery 13 │ 38.71 / 39.86 ±1.38 / 42.50 ms │ 38.61 / 38.73 ±0.12 / 38.90 ms │     no change │
│ QQuery 14 │ 24.47 / 25.23 ±1.13 / 27.43 ms │ 24.34 / 24.81 ±0.62 / 26.03 ms │     no change │
│ QQuery 15 │ 31.13 / 31.88 ±0.65 / 33.07 ms │ 30.46 / 31.22 ±0.83 / 32.65 ms │     no change │
│ QQuery 16 │ 13.52 / 13.72 ±0.17 / 13.99 ms │ 13.40 / 13.58 ±0.18 / 13.92 ms │     no change │
│ QQuery 17 │ 73.28 / 76.26 ±4.09 / 84.28 ms │ 73.17 / 73.75 ±0.53 / 74.51 ms │     no change │
│ QQuery 18 │ 44.71 / 46.47 ±1.52 / 48.92 ms │ 45.31 / 45.84 ±0.63 / 47.01 ms │     no change │
│ QQuery 19 │ 31.59 / 32.64 ±1.66 / 35.91 ms │ 31.74 / 32.55 ±0.65 / 33.36 ms │     no change │
│ QQuery 20 │ 31.45 / 32.30 ±0.73 / 33.61 ms │ 31.86 / 32.63 ±0.69 / 33.50 ms │     no change │
│ QQuery 21 │ 55.46 / 56.44 ±1.27 / 58.90 ms │ 54.91 / 56.73 ±2.28 / 60.83 ms │     no change │
│ QQuery 22 │ 13.82 / 14.73 ±1.22 / 17.11 ms │ 13.34 / 13.67 ±0.19 / 13.90 ms │ +1.08x faster │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary              ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)              │ 727.98ms │
│ Total Time (sort-background)   │ 722.28ms │
│ Average Time (HEAD)            │  33.09ms │
│ Average Time (sort-background) │  32.83ms │
│ Queries Faster                 │        1 │
│ Queries Slower                 │        0 │
│ Queries with No Change         │       21 │
│ Queries with Failure           │        0 │
└────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 475.6 MiB
CPU user 19.8s
CPU sys 1.6s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.2 GiB
Avg memory 478.8 MiB
CPU user 19.7s
CPU sys 1.6s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort_tpch
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and sort-background
--------------------
Benchmark sort_tpch1.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃      HEAD ┃ sort-background ┃       Change ┃
┡━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ Q1    │ 110.76 ms │       118.66 ms │ 1.07x slower │
│ Q2    │  96.62 ms │       115.69 ms │ 1.20x slower │
│ Q3    │ 637.45 ms │       632.64 ms │    no change │
│ Q4    │ 159.16 ms │       175.18 ms │ 1.10x slower │
│ Q5    │ 233.24 ms │       239.25 ms │    no change │
│ Q6    │ 245.86 ms │       250.67 ms │    no change │
│ Q7    │ 409.70 ms │       409.84 ms │    no change │
│ Q8    │ 294.31 ms │       294.40 ms │    no change │
│ Q9    │ 314.02 ms │       309.29 ms │    no change │
│ Q10   │ 458.09 ms │       493.02 ms │ 1.08x slower │
│ Q11   │ 246.65 ms │       244.37 ms │    no change │
└───────┴───────────┴─────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 3205.86ms │
│ Total Time (sort-background)   │ 3283.03ms │
│ Average Time (HEAD)            │  291.44ms │
│ Average Time (sort-background) │  298.46ms │
│ Queries Faster                 │         0 │
│ Queries Slower                 │         4 │
│ Queries with No Change         │         7 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and sort-background
--------------------
Benchmark sort_tpch1.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃                               HEAD ┃                    sort-background ┃       Change ┃
┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ Q1    │  110.76 / 111.64 ±1.01 / 113.55 ms │  118.66 / 119.62 ±0.78 / 120.92 ms │ 1.07x slower │
│ Q2    │    96.62 / 97.69 ±1.49 / 100.60 ms │  115.69 / 117.62 ±1.14 / 118.82 ms │ 1.20x slower │
│ Q3    │  637.45 / 639.92 ±1.44 / 641.82 ms │  632.64 / 636.30 ±2.72 / 641.00 ms │    no change │
│ Q4    │  159.16 / 159.96 ±0.69 / 161.22 ms │  175.18 / 179.64 ±6.16 / 191.77 ms │ 1.12x slower │
│ Q5    │  233.24 / 234.62 ±1.11 / 236.56 ms │  239.25 / 240.59 ±1.11 / 242.48 ms │    no change │
│ Q6    │  245.86 / 247.47 ±1.62 / 249.93 ms │  250.67 / 251.98 ±1.71 / 255.32 ms │    no change │
│ Q7    │  409.70 / 412.37 ±2.38 / 416.76 ms │  409.84 / 411.15 ±0.89 / 412.37 ms │    no change │
│ Q8    │  294.31 / 300.02 ±5.40 / 306.88 ms │  294.40 / 302.33 ±7.90 / 316.95 ms │    no change │
│ Q9    │ 314.02 / 332.41 ±11.58 / 345.56 ms │  309.29 / 316.00 ±6.57 / 328.50 ms │    no change │
│ Q10   │ 458.09 / 483.54 ±25.62 / 529.74 ms │ 493.02 / 502.81 ±10.11 / 517.48 ms │    no change │
│ Q11   │ 246.65 / 261.56 ±25.03 / 311.43 ms │ 244.37 / 256.86 ±12.63 / 280.48 ms │    no change │
└───────┴────────────────────────────────────┴────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 3281.19ms │
│ Total Time (sort-background)   │ 3334.89ms │
│ Average Time (HEAD)            │  298.29ms │
│ Average Time (sort-background) │  303.17ms │
│ Queries Faster                 │         0 │
│ Queries Slower                 │         3 │
│ Queries with No Change         │         8 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Resource Usage

sort_tpch — base (merge-base)

Metric Value
Wall time 20.0s
Peak memory 2.4 GiB
Avg memory 1.0 GiB
CPU user 60.6s
CPU sys 3.2s
Peak spill 0 B

sort_tpch — branch

Metric Value
Wall time 20.0s
Peak memory 2.4 GiB
Avg memory 1.0 GiB
CPU user 60.7s
CPU sys 3.2s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpch10
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and sort-background
--------------------
Benchmark tpch_sf10.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃      HEAD ┃ sort-background ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 333.63 ms │       341.07 ms │     no change │
│ QQuery 2  │  92.63 ms │        91.04 ms │     no change │
│ QQuery 3  │ 221.48 ms │       230.27 ms │     no change │
│ QQuery 4  │ 114.09 ms │       124.41 ms │  1.09x slower │
│ QQuery 5  │ 358.17 ms │       367.19 ms │     no change │
│ QQuery 6  │ 105.30 ms │       103.48 ms │     no change │
│ QQuery 7  │ 463.23 ms │       485.68 ms │     no change │
│ QQuery 8  │ 360.84 ms │       374.96 ms │     no change │
│ QQuery 9  │ 523.69 ms │       510.54 ms │     no change │
│ QQuery 10 │ 305.65 ms │       291.78 ms │     no change │
│ QQuery 11 │  62.08 ms │        62.23 ms │     no change │
│ QQuery 12 │ 120.85 ms │       120.28 ms │     no change │
│ QQuery 13 │ 308.23 ms │       303.24 ms │     no change │
│ QQuery 14 │ 171.65 ms │       174.12 ms │     no change │
│ QQuery 15 │ 302.47 ms │       306.40 ms │     no change │
│ QQuery 16 │  64.94 ms │        63.79 ms │     no change │
│ QQuery 17 │ 601.60 ms │       585.87 ms │     no change │
│ QQuery 18 │ 767.22 ms │       745.24 ms │     no change │
│ QQuery 19 │ 247.00 ms │       237.37 ms │     no change │
│ QQuery 20 │ 291.16 ms │       271.03 ms │ +1.07x faster │
│ QQuery 21 │ 736.45 ms │       668.22 ms │ +1.10x faster │
│ QQuery 22 │  62.78 ms │        59.23 ms │ +1.06x faster │
└───────────┴───────────┴─────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 6615.15ms │
│ Total Time (sort-background)   │ 6517.46ms │
│ Average Time (HEAD)            │  300.69ms │
│ Average Time (sort-background) │  296.25ms │
│ Queries Faster                 │         3 │
│ Queries Slower                 │         1 │
│ Queries with No Change         │        18 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and sort-background
--------------------
Benchmark tpch_sf10.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                               HEAD ┃                    sort-background ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │  333.63 / 341.63 ±4.19 / 345.69 ms │  341.07 / 346.49 ±3.24 / 350.09 ms │     no change │
│ QQuery 2  │     92.63 / 95.66 ±2.13 / 97.97 ms │     91.04 / 94.82 ±3.31 / 99.40 ms │     no change │
│ QQuery 3  │  221.48 / 225.93 ±4.62 / 233.77 ms │  230.27 / 241.28 ±6.16 / 247.88 ms │  1.07x slower │
│ QQuery 4  │  114.09 / 118.89 ±3.30 / 123.47 ms │  124.41 / 126.30 ±1.29 / 127.86 ms │  1.06x slower │
│ QQuery 5  │ 358.17 / 375.07 ±11.14 / 389.52 ms │ 367.19 / 382.81 ±14.36 / 401.49 ms │     no change │
│ QQuery 6  │  105.30 / 109.11 ±4.07 / 116.83 ms │  103.48 / 109.36 ±3.20 / 113.07 ms │     no change │
│ QQuery 7  │  463.23 / 471.60 ±8.95 / 486.39 ms │ 485.68 / 501.27 ±12.94 / 518.91 ms │  1.06x slower │
│ QQuery 8  │  360.84 / 365.93 ±4.02 / 369.98 ms │  374.96 / 379.79 ±3.46 / 383.79 ms │     no change │
│ QQuery 9  │ 523.69 / 551.95 ±18.64 / 580.83 ms │ 510.54 / 541.01 ±19.88 / 568.90 ms │     no change │
│ QQuery 10 │  305.65 / 311.85 ±4.99 / 320.20 ms │  291.78 / 304.08 ±9.73 / 319.38 ms │     no change │
│ QQuery 11 │     62.08 / 66.02 ±5.99 / 77.93 ms │     62.23 / 63.92 ±1.21 / 65.91 ms │     no change │
│ QQuery 12 │  120.85 / 125.49 ±4.54 / 132.51 ms │ 120.28 / 129.22 ±10.48 / 148.60 ms │     no change │
│ QQuery 13 │  308.23 / 318.89 ±9.59 / 331.77 ms │ 303.24 / 322.70 ±15.62 / 342.03 ms │     no change │
│ QQuery 14 │  171.65 / 177.07 ±5.94 / 185.30 ms │  174.12 / 181.75 ±6.58 / 191.85 ms │     no change │
│ QQuery 15 │  302.47 / 310.00 ±4.40 / 314.61 ms │  306.40 / 313.22 ±5.17 / 320.64 ms │     no change │
│ QQuery 16 │     64.94 / 66.53 ±1.98 / 70.40 ms │     63.79 / 65.12 ±0.90 / 66.16 ms │     no change │
│ QQuery 17 │ 601.60 / 616.34 ±16.75 / 645.09 ms │ 585.87 / 621.34 ±30.66 / 664.59 ms │     no change │
│ QQuery 18 │ 767.22 / 806.45 ±46.84 / 886.54 ms │ 745.24 / 797.62 ±31.64 / 832.18 ms │     no change │
│ QQuery 19 │ 247.00 / 264.42 ±24.18 / 311.84 ms │  237.37 / 250.87 ±9.50 / 262.56 ms │ +1.05x faster │
│ QQuery 20 │  291.16 / 300.95 ±6.73 / 309.57 ms │ 271.03 / 283.67 ±11.86 / 304.64 ms │ +1.06x faster │
│ QQuery 21 │ 736.45 / 759.01 ±13.69 / 776.36 ms │ 668.22 / 704.88 ±22.63 / 732.34 ms │ +1.08x faster │
│ QQuery 22 │     62.78 / 65.07 ±2.09 / 68.77 ms │     59.23 / 64.71 ±4.54 / 70.43 ms │     no change │
└───────────┴────────────────────────────────────┴────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 6843.87ms │
│ Total Time (sort-background)   │ 6826.23ms │
│ Average Time (HEAD)            │  311.08ms │
│ Average Time (sort-background) │  310.28ms │
│ Queries Faster                 │         3 │
│ Queries Slower                 │         3 │
│ Queries with No Change         │        16 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Resource Usage

tpch10 — base (merge-base)

Metric Value
Wall time 35.0s
Peak memory 5.1 GiB
Avg memory 1.6 GiB
CPU user 345.1s
CPU sys 21.3s
Peak spill 0 B

tpch10 — branch

Metric Value
Wall time 35.0s
Peak memory 5.1 GiB
Avg memory 1.5 GiB
CPU user 346.6s
CPU sys 20.9s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpcds
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and sort-background
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ sort-background ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.04 ms │         5.35 ms │  1.06x slower │
│ QQuery 2  │   80.95 ms │        81.03 ms │     no change │
│ QQuery 3  │   28.88 ms │        28.83 ms │     no change │
│ QQuery 4  │  467.93 ms │       458.19 ms │     no change │
│ QQuery 5  │   49.89 ms │        49.36 ms │     no change │
│ QQuery 6  │   34.93 ms │        34.90 ms │     no change │
│ QQuery 7  │   71.63 ms │        71.60 ms │     no change │
│ QQuery 8  │   35.72 ms │        35.24 ms │     no change │
│ QQuery 9  │   51.42 ms │        49.59 ms │     no change │
│ QQuery 10 │   59.65 ms │        60.38 ms │     no change │
│ QQuery 11 │  270.54 ms │       270.51 ms │     no change │
│ QQuery 12 │   28.52 ms │        28.48 ms │     no change │
│ QQuery 13 │  114.32 ms │       112.60 ms │     no change │
│ QQuery 14 │  384.49 ms │       386.03 ms │     no change │
│ QQuery 15 │   55.32 ms │        55.03 ms │     no change │
│ QQuery 16 │    6.13 ms │         5.97 ms │     no change │
│ QQuery 17 │   78.05 ms │        78.11 ms │     no change │
│ QQuery 18 │  104.42 ms │       103.97 ms │     no change │
│ QQuery 19 │   40.83 ms │        40.63 ms │     no change │
│ QQuery 20 │   34.90 ms │        35.70 ms │     no change │
│ QQuery 21 │   16.62 ms │        17.14 ms │     no change │
│ QQuery 22 │   64.65 ms │        64.82 ms │     no change │
│ QQuery 23 │  311.45 ms │       309.11 ms │     no change │
│ QQuery 24 │  192.46 ms │       197.41 ms │     no change │
│ QQuery 25 │  108.14 ms │       105.73 ms │     no change │
│ QQuery 26 │   45.53 ms │        45.72 ms │     no change │
│ QQuery 27 │    6.06 ms │         6.00 ms │     no change │
│ QQuery 28 │   56.37 ms │        59.38 ms │  1.05x slower │
│ QQuery 29 │   93.84 ms │        93.83 ms │     no change │
│ QQuery 30 │   32.24 ms │        32.45 ms │     no change │
│ QQuery 31 │  108.69 ms │       107.81 ms │     no change │
│ QQuery 32 │   20.35 ms │        20.07 ms │     no change │
│ QQuery 33 │   36.42 ms │        35.92 ms │     no change │
│ QQuery 34 │   10.06 ms │         9.99 ms │     no change │
│ QQuery 35 │   72.69 ms │        72.60 ms │     no change │
│ QQuery 36 │    5.81 ms │         5.76 ms │     no change │
│ QQuery 37 │    6.89 ms │         6.81 ms │     no change │
│ QQuery 38 │   62.48 ms │        61.79 ms │     no change │
│ QQuery 39 │   72.05 ms │        73.55 ms │     no change │
│ QQuery 40 │   23.68 ms │        23.13 ms │     no change │
│ QQuery 41 │   12.09 ms │        12.10 ms │     no change │
│ QQuery 42 │   23.74 ms │        23.49 ms │     no change │
│ QQuery 43 │    4.84 ms │         4.55 ms │ +1.06x faster │
│ QQuery 44 │    8.76 ms │         8.28 ms │ +1.06x faster │
│ QQuery 45 │   36.29 ms │        36.32 ms │     no change │
│ QQuery 46 │   11.26 ms │        11.23 ms │     no change │
│ QQuery 47 │  223.72 ms │       226.57 ms │     no change │
│ QQuery 48 │   89.28 ms │        90.51 ms │     no change │
│ QQuery 49 │   68.34 ms │        68.86 ms │     no change │
│ QQuery 50 │   58.08 ms │        57.69 ms │     no change │
│ QQuery 51 │   92.24 ms │        86.66 ms │ +1.06x faster │
│ QQuery 52 │   24.23 ms │        24.57 ms │     no change │
│ QQuery 53 │   29.04 ms │        29.10 ms │     no change │
│ QQuery 54 │   52.91 ms │        53.14 ms │     no change │
│ QQuery 55 │   23.75 ms │        23.40 ms │     no change │
│ QQuery 56 │   37.69 ms │        37.16 ms │     no change │
│ QQuery 57 │  166.03 ms │       163.81 ms │     no change │
│ QQuery 58 │  107.74 ms │       106.48 ms │     no change │
│ QQuery 59 │  115.31 ms │       115.75 ms │     no change │
│ QQuery 60 │   37.75 ms │        38.66 ms │     no change │
│ QQuery 61 │   10.38 ms │        10.75 ms │     no change │
│ QQuery 62 │   43.74 ms │        44.34 ms │     no change │
│ QQuery 63 │   29.34 ms │        29.27 ms │     no change │
│ QQuery 64 │  352.86 ms │       359.44 ms │     no change │
│ QQuery 65 │  130.78 ms │       128.86 ms │     no change │
│ QQuery 66 │   73.88 ms │        75.72 ms │     no change │
│ QQuery 67 │  237.25 ms │       256.74 ms │  1.08x slower │
│ QQuery 68 │   11.18 ms │        11.72 ms │     no change │
│ QQuery 69 │   54.49 ms │        55.74 ms │     no change │
│ QQuery 70 │  104.53 ms │       104.36 ms │     no change │
│ QQuery 71 │   33.97 ms │        34.63 ms │     no change │
│ QQuery 72 │ 1562.13 ms │      1585.55 ms │     no change │
│ QQuery 73 │    9.82 ms │         9.86 ms │     no change │
│ QQuery 74 │  151.04 ms │       157.46 ms │     no change │
│ QQuery 75 │  138.21 ms │       139.44 ms │     no change │
│ QQuery 76 │   34.42 ms │        34.12 ms │     no change │
│ QQuery 77 │   59.28 ms │        59.28 ms │     no change │
│ QQuery 78 │  160.90 ms │       160.30 ms │     no change │
│ QQuery 79 │   65.30 ms │        65.66 ms │     no change │
│ QQuery 80 │   94.16 ms │        93.33 ms │     no change │
│ QQuery 81 │   25.76 ms │        25.50 ms │     no change │
│ QQuery 82 │   16.68 ms │        16.60 ms │     no change │
│ QQuery 83 │   32.88 ms │        33.17 ms │     no change │
│ QQuery 84 │   30.42 ms │        29.78 ms │     no change │
│ QQuery 85 │   98.48 ms │        97.79 ms │     no change │
│ QQuery 86 │   25.11 ms │        25.86 ms │     no change │
│ QQuery 87 │   59.84 ms │        62.16 ms │     no change │
│ QQuery 88 │   58.27 ms │        58.49 ms │     no change │
│ QQuery 89 │   35.02 ms │        34.88 ms │     no change │
│ QQuery 90 │   16.34 ms │        16.11 ms │     no change │
│ QQuery 91 │   40.90 ms │        41.73 ms │     no change │
│ QQuery 92 │   29.65 ms │        29.30 ms │     no change │
│ QQuery 93 │   48.30 ms │        49.70 ms │     no change │
│ QQuery 94 │   38.52 ms │        38.20 ms │     no change │
│ QQuery 95 │   79.45 ms │        79.25 ms │     no change │
│ QQuery 96 │   23.76 ms │        23.62 ms │     no change │
│ QQuery 97 │   50.56 ms │        49.66 ms │     no change │
│ QQuery 98 │   42.05 ms │        42.23 ms │     no change │
│ QQuery 99 │   64.95 ms │        64.91 ms │     no change │
└───────────┴────────────┴─────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 8747.70ms │
│ Total Time (sort-background)   │ 8794.34ms │
│ Average Time (HEAD)            │   88.36ms │
│ Average Time (sort-background) │   88.83ms │
│ Queries Faster                 │         3 │
│ Queries Slower                 │         3 │
│ Queries with No Change         │        93 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and sort-background
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                       sort-background ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.04 / 5.71 ±1.07 / 7.83 ms │           5.35 / 6.08 ±0.87 / 7.72 ms │  1.07x slower │
│ QQuery 2  │        80.95 / 81.24 ±0.19 / 81.53 ms │        81.03 / 81.41 ±0.24 / 81.71 ms │     no change │
│ QQuery 3  │        28.88 / 29.07 ±0.21 / 29.45 ms │        28.83 / 29.03 ±0.17 / 29.29 ms │     no change │
│ QQuery 4  │    467.93 / 477.01 ±11.66 / 499.99 ms │     458.19 / 466.65 ±6.58 / 474.84 ms │     no change │
│ QQuery 5  │        49.89 / 52.81 ±4.09 / 60.76 ms │        49.36 / 50.00 ±0.68 / 51.14 ms │ +1.06x faster │
│ QQuery 6  │        34.93 / 35.31 ±0.29 / 35.82 ms │        34.90 / 35.16 ±0.23 / 35.56 ms │     no change │
│ QQuery 7  │        71.63 / 72.13 ±0.44 / 72.93 ms │        71.60 / 72.26 ±0.60 / 73.27 ms │     no change │
│ QQuery 8  │        35.72 / 36.07 ±0.21 / 36.31 ms │        35.24 / 35.76 ±0.39 / 36.31 ms │     no change │
│ QQuery 9  │        51.42 / 53.04 ±1.04 / 54.33 ms │        49.59 / 53.80 ±2.29 / 56.03 ms │     no change │
│ QQuery 10 │        59.65 / 60.34 ±0.44 / 60.96 ms │        60.38 / 60.78 ±0.25 / 61.04 ms │     no change │
│ QQuery 11 │    270.54 / 288.89 ±11.68 / 304.29 ms │     270.51 / 276.88 ±7.08 / 288.80 ms │     no change │
│ QQuery 12 │        28.52 / 30.45 ±2.92 / 36.24 ms │        28.48 / 28.72 ±0.16 / 28.96 ms │ +1.06x faster │
│ QQuery 13 │     114.32 / 115.39 ±0.74 / 116.16 ms │     112.60 / 114.73 ±1.72 / 117.53 ms │     no change │
│ QQuery 14 │     384.49 / 387.50 ±4.03 / 395.33 ms │     386.03 / 388.15 ±1.83 / 391.21 ms │     no change │
│ QQuery 15 │        55.32 / 56.24 ±0.63 / 57.25 ms │        55.03 / 56.91 ±2.21 / 60.98 ms │     no change │
│ QQuery 16 │           6.13 / 6.34 ±0.22 / 6.78 ms │           5.97 / 6.14 ±0.17 / 6.46 ms │     no change │
│ QQuery 17 │        78.05 / 80.15 ±2.31 / 84.41 ms │        78.11 / 79.05 ±0.84 / 80.57 ms │     no change │
│ QQuery 18 │     104.42 / 105.93 ±1.53 / 107.94 ms │     103.97 / 105.24 ±1.37 / 107.61 ms │     no change │
│ QQuery 19 │        40.83 / 41.72 ±1.12 / 43.83 ms │        40.63 / 41.18 ±0.40 / 41.66 ms │     no change │
│ QQuery 20 │        34.90 / 35.21 ±0.16 / 35.34 ms │        35.70 / 36.05 ±0.29 / 36.34 ms │     no change │
│ QQuery 21 │        16.62 / 16.88 ±0.21 / 17.17 ms │        17.14 / 17.34 ±0.19 / 17.63 ms │     no change │
│ QQuery 22 │        64.65 / 65.80 ±1.08 / 67.27 ms │        64.82 / 65.71 ±0.52 / 66.28 ms │     no change │
│ QQuery 23 │     311.45 / 314.76 ±2.50 / 318.91 ms │     309.11 / 314.82 ±4.18 / 319.80 ms │     no change │
│ QQuery 24 │     192.46 / 199.51 ±6.13 / 208.52 ms │     197.41 / 200.36 ±2.98 / 204.96 ms │     no change │
│ QQuery 25 │     108.14 / 108.73 ±0.34 / 109.01 ms │     105.73 / 107.61 ±1.03 / 108.80 ms │     no change │
│ QQuery 26 │        45.53 / 46.13 ±0.39 / 46.72 ms │        45.72 / 46.37 ±0.62 / 47.52 ms │     no change │
│ QQuery 27 │           6.06 / 6.14 ±0.05 / 6.19 ms │           6.00 / 6.13 ±0.14 / 6.40 ms │     no change │
│ QQuery 28 │        56.37 / 59.76 ±3.01 / 64.45 ms │        59.38 / 60.34 ±0.53 / 60.82 ms │     no change │
│ QQuery 29 │        93.84 / 97.31 ±2.06 / 99.84 ms │       93.83 / 97.86 ±5.16 / 107.90 ms │     no change │
│ QQuery 30 │        32.24 / 33.10 ±0.68 / 34.18 ms │        32.45 / 33.11 ±0.46 / 33.81 ms │     no change │
│ QQuery 31 │     108.69 / 109.43 ±0.68 / 110.55 ms │     107.81 / 108.58 ±0.84 / 109.99 ms │     no change │
│ QQuery 32 │        20.35 / 20.77 ±0.25 / 21.03 ms │        20.07 / 20.63 ±0.67 / 21.94 ms │     no change │
│ QQuery 33 │        36.42 / 38.27 ±2.41 / 42.91 ms │        35.92 / 36.69 ±0.58 / 37.68 ms │     no change │
│ QQuery 34 │        10.06 / 10.21 ±0.13 / 10.45 ms │         9.99 / 10.38 ±0.40 / 11.15 ms │     no change │
│ QQuery 35 │        72.69 / 74.57 ±1.72 / 77.77 ms │        72.60 / 74.14 ±1.52 / 76.28 ms │     no change │
│ QQuery 36 │           5.81 / 5.99 ±0.12 / 6.17 ms │           5.76 / 5.94 ±0.12 / 6.10 ms │     no change │
│ QQuery 37 │           6.89 / 7.03 ±0.19 / 7.41 ms │           6.81 / 6.98 ±0.11 / 7.11 ms │     no change │
│ QQuery 38 │        62.48 / 62.76 ±0.40 / 63.54 ms │        61.79 / 62.38 ±0.57 / 63.35 ms │     no change │
│ QQuery 39 │        72.05 / 73.54 ±1.23 / 75.60 ms │        73.55 / 75.32 ±2.94 / 81.13 ms │     no change │
│ QQuery 40 │        23.68 / 24.54 ±1.19 / 26.87 ms │        23.13 / 23.84 ±0.50 / 24.45 ms │     no change │
│ QQuery 41 │        12.09 / 12.24 ±0.14 / 12.45 ms │        12.10 / 12.32 ±0.21 / 12.67 ms │     no change │
│ QQuery 42 │        23.74 / 24.15 ±0.23 / 24.41 ms │        23.49 / 24.55 ±0.69 / 25.31 ms │     no change │
│ QQuery 43 │           4.84 / 4.95 ±0.14 / 5.22 ms │           4.55 / 4.69 ±0.23 / 5.15 ms │ +1.05x faster │
│ QQuery 44 │           8.76 / 8.86 ±0.07 / 8.96 ms │           8.28 / 8.44 ±0.15 / 8.72 ms │     no change │
│ QQuery 45 │        36.29 / 36.96 ±0.55 / 37.77 ms │        36.32 / 36.74 ±0.46 / 37.64 ms │     no change │
│ QQuery 46 │        11.26 / 11.43 ±0.17 / 11.69 ms │        11.23 / 11.40 ±0.14 / 11.63 ms │     no change │
│ QQuery 47 │     223.72 / 226.53 ±3.93 / 234.00 ms │     226.57 / 230.15 ±2.80 / 232.83 ms │     no change │
│ QQuery 48 │        89.28 / 90.79 ±1.17 / 92.86 ms │        90.51 / 91.26 ±0.63 / 92.20 ms │     no change │
│ QQuery 49 │        68.34 / 69.11 ±0.56 / 70.02 ms │        68.86 / 69.24 ±0.36 / 69.83 ms │     no change │
│ QQuery 50 │        58.08 / 59.47 ±2.29 / 63.99 ms │        57.69 / 58.90 ±1.24 / 61.26 ms │     no change │
│ QQuery 51 │        92.24 / 93.61 ±1.13 / 95.00 ms │        86.66 / 89.20 ±1.57 / 91.40 ms │     no change │
│ QQuery 52 │        24.23 / 24.34 ±0.09 / 24.48 ms │        24.57 / 24.77 ±0.20 / 25.14 ms │     no change │
│ QQuery 53 │        29.04 / 29.34 ±0.18 / 29.60 ms │        29.10 / 29.33 ±0.21 / 29.72 ms │     no change │
│ QQuery 54 │        52.91 / 54.60 ±1.55 / 57.34 ms │        53.14 / 54.43 ±1.99 / 58.35 ms │     no change │
│ QQuery 55 │        23.75 / 23.96 ±0.15 / 24.20 ms │        23.40 / 23.71 ±0.36 / 24.29 ms │     no change │
│ QQuery 56 │        37.69 / 38.60 ±1.08 / 40.72 ms │        37.16 / 37.48 ±0.25 / 37.78 ms │     no change │
│ QQuery 57 │     166.03 / 166.90 ±0.96 / 168.74 ms │     163.81 / 166.68 ±3.18 / 172.74 ms │     no change │
│ QQuery 58 │     107.74 / 108.34 ±0.71 / 109.74 ms │     106.48 / 108.16 ±1.22 / 109.49 ms │     no change │
│ QQuery 59 │     115.31 / 117.00 ±1.75 / 119.94 ms │     115.75 / 117.28 ±2.21 / 121.65 ms │     no change │
│ QQuery 60 │        37.75 / 38.38 ±0.85 / 40.04 ms │        38.66 / 38.94 ±0.21 / 39.32 ms │     no change │
│ QQuery 61 │        10.38 / 10.46 ±0.08 / 10.60 ms │        10.75 / 10.91 ±0.13 / 11.15 ms │     no change │
│ QQuery 62 │        43.74 / 44.15 ±0.36 / 44.57 ms │        44.34 / 44.81 ±0.27 / 45.12 ms │     no change │
│ QQuery 63 │        29.34 / 30.08 ±0.52 / 30.95 ms │        29.27 / 29.49 ±0.13 / 29.66 ms │     no change │
│ QQuery 64 │     352.86 / 357.83 ±3.03 / 361.06 ms │     359.44 / 363.99 ±3.59 / 370.17 ms │     no change │
│ QQuery 65 │     130.78 / 132.62 ±1.74 / 135.83 ms │     128.86 / 130.32 ±0.74 / 130.81 ms │     no change │
│ QQuery 66 │        73.88 / 74.51 ±0.59 / 75.30 ms │        75.72 / 76.30 ±0.41 / 76.88 ms │     no change │
│ QQuery 67 │     237.25 / 241.38 ±4.24 / 248.57 ms │     256.74 / 260.10 ±2.50 / 263.83 ms │  1.08x slower │
│ QQuery 68 │        11.18 / 11.36 ±0.19 / 11.74 ms │        11.72 / 11.88 ±0.20 / 12.27 ms │     no change │
│ QQuery 69 │        54.49 / 55.66 ±1.07 / 57.61 ms │        55.74 / 57.22 ±2.46 / 62.11 ms │     no change │
│ QQuery 70 │     104.53 / 105.27 ±0.96 / 107.17 ms │     104.36 / 105.47 ±1.26 / 107.91 ms │     no change │
│ QQuery 71 │        33.97 / 36.43 ±2.80 / 40.31 ms │        34.63 / 36.72 ±2.77 / 42.10 ms │     no change │
│ QQuery 72 │ 1562.13 / 1641.25 ±72.58 / 1758.44 ms │ 1585.55 / 1628.72 ±31.96 / 1678.93 ms │     no change │
│ QQuery 73 │         9.82 / 10.09 ±0.24 / 10.46 ms │         9.86 / 10.31 ±0.34 / 10.70 ms │     no change │
│ QQuery 74 │     151.04 / 156.41 ±3.41 / 159.91 ms │     157.46 / 164.02 ±4.99 / 170.68 ms │     no change │
│ QQuery 75 │     138.21 / 139.47 ±1.42 / 142.12 ms │     139.44 / 141.03 ±1.11 / 142.12 ms │     no change │
│ QQuery 76 │        34.42 / 36.24 ±2.54 / 41.24 ms │        34.12 / 36.23 ±3.82 / 43.85 ms │     no change │
│ QQuery 77 │        59.28 / 60.35 ±0.77 / 61.63 ms │        59.28 / 59.95 ±0.47 / 60.70 ms │     no change │
│ QQuery 78 │     160.90 / 168.94 ±5.70 / 178.46 ms │     160.30 / 166.21 ±5.43 / 172.76 ms │     no change │
│ QQuery 79 │        65.30 / 66.62 ±0.86 / 67.70 ms │        65.66 / 66.68 ±0.71 / 67.69 ms │     no change │
│ QQuery 80 │        94.16 / 95.67 ±1.34 / 97.49 ms │        93.33 / 95.47 ±1.88 / 98.88 ms │     no change │
│ QQuery 81 │        25.76 / 27.32 ±2.24 / 31.75 ms │        25.50 / 27.34 ±2.64 / 32.49 ms │     no change │
│ QQuery 82 │        16.68 / 16.80 ±0.13 / 16.97 ms │        16.60 / 18.47 ±2.94 / 24.32 ms │  1.10x slower │
│ QQuery 83 │        32.88 / 33.08 ±0.18 / 33.36 ms │        33.17 / 33.38 ±0.21 / 33.67 ms │     no change │
│ QQuery 84 │        30.42 / 30.79 ±0.26 / 31.17 ms │        29.78 / 30.27 ±0.56 / 30.96 ms │     no change │
│ QQuery 85 │      98.48 / 100.74 ±2.47 / 105.48 ms │       97.79 / 98.88 ±1.29 / 101.34 ms │     no change │
│ QQuery 86 │        25.11 / 25.70 ±0.59 / 26.69 ms │        25.86 / 26.50 ±0.43 / 26.96 ms │     no change │
│ QQuery 87 │        59.84 / 61.99 ±1.45 / 63.78 ms │        62.16 / 64.27 ±2.72 / 69.48 ms │     no change │
│ QQuery 88 │        58.27 / 59.05 ±0.57 / 59.83 ms │        58.49 / 59.20 ±0.49 / 59.71 ms │     no change │
│ QQuery 89 │        35.02 / 35.57 ±0.55 / 36.42 ms │        34.88 / 35.29 ±0.38 / 35.98 ms │     no change │
│ QQuery 90 │        16.34 / 17.92 ±2.93 / 23.78 ms │        16.11 / 17.26 ±1.96 / 21.17 ms │     no change │
│ QQuery 91 │        40.90 / 41.80 ±0.86 / 42.91 ms │        41.73 / 42.95 ±1.20 / 44.97 ms │     no change │
│ QQuery 92 │        29.65 / 30.34 ±0.38 / 30.77 ms │        29.30 / 29.74 ±0.26 / 30.08 ms │     no change │
│ QQuery 93 │        48.30 / 49.67 ±0.93 / 51.06 ms │        49.70 / 50.03 ±0.26 / 50.43 ms │     no change │
│ QQuery 94 │        38.52 / 38.80 ±0.24 / 39.19 ms │        38.20 / 38.46 ±0.21 / 38.70 ms │     no change │
│ QQuery 95 │        79.45 / 81.12 ±1.58 / 84.05 ms │        79.25 / 81.24 ±2.02 / 84.82 ms │     no change │
│ QQuery 96 │        23.76 / 23.88 ±0.10 / 24.06 ms │        23.62 / 23.91 ±0.21 / 24.17 ms │     no change │
│ QQuery 97 │        50.56 / 52.11 ±1.62 / 54.99 ms │        49.66 / 51.34 ±1.59 / 54.16 ms │     no change │
│ QQuery 98 │        42.05 / 42.62 ±0.53 / 43.43 ms │        42.23 / 42.71 ±0.42 / 43.41 ms │     no change │
│ QQuery 99 │        64.95 / 66.30 ±1.68 / 69.58 ms │        64.91 / 66.21 ±2.03 / 70.26 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 8975.72ms │
│ Total Time (sort-background)   │ 8969.52ms │
│ Average Time (HEAD)            │   90.66ms │
│ Average Time (sort-background) │   90.60ms │
│ Queries Faster                 │         3 │
│ Queries Slower                 │         3 │
│ Queries with No Change         │        93 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 50.0s
Peak memory 2.3 GiB
Avg memory 1.5 GiB
CPU user 190.6s
CPU sys 5.8s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 50.0s
Peak memory 2.4 GiB
Avg memory 1.5 GiB
CPU user 189.8s
CPU sys 5.7s
Peak spill 0 B

File an issue against this benchmark runner

@rluvaton

rluvaton commented Oct 20, 2026

Copy link
Copy Markdown
Member Author

run benchmarks sort_tpch10

env:
   prefer_hash_join: false

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6291792333-3637-7b4pg 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort_tpch10
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort
env:
  BENCH_FILTER: "sort_axis/sort.*.1M.High.cardinality.Unsorted..100cols"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

group                                                               HEAD                                   sort-background
-----                                                               ----                                   ---------------
sort_axis/sort i64 1M High cardinality Unsorted +100cols            1.00       2.2±0.07s        ? ?/sec    1.02       2.2±0.04s        ? ?/sec
sort_axis/sort mixed tuple 1M High cardinality Unsorted +100cols    1.03       2.4±0.03s        ? ?/sec    1.00       2.3±0.04s        ? ?/sec
sort_axis/sort utf8 view 1M High cardinality Unsorted +100cols      1.00       2.2±0.08s        ? ?/sec    1.00       2.2±0.04s        ? ?/sec

Resource Usage

sort — base (merge-base)

Metric Value
Wall time 2615.6s
Peak memory 2.3 GiB
Avg memory 1.3 GiB
CPU user 2224.0s
CPU sys 115.2s
Peak spill 0 B

sort — branch

Metric Value
Wall time 2565.6s
Peak memory 2.3 GiB
Avg memory 1.3 GiB
CPU user 2210.2s
CPU sys 116.4s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (a60b142) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort_tpch10
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and sort-background
--------------------
Benchmark sort_tpch10.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃       HEAD ┃ sort-background ┃       Change ┃
┡━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ Q1    │ 1403.13 ms │      1697.90 ms │ 1.21x slower │
│ Q2    │ 1281.30 ms │      1670.01 ms │ 1.30x slower │
│ Q3    │ 8521.15 ms │      8471.71 ms │    no change │
│ Q4    │ 2209.82 ms │      2607.15 ms │ 1.18x slower │
│ Q5    │ 2391.88 ms │      2416.08 ms │    no change │
│ Q6    │ 2544.64 ms │      2568.14 ms │    no change │
│ Q7    │ 4972.19 ms │      5050.45 ms │    no change │
│ Q8    │ 3454.39 ms │      3772.64 ms │ 1.09x slower │
│ Q9    │ 3833.18 ms │      4071.69 ms │ 1.06x slower │
│ Q10   │ 6038.40 ms │      6289.16 ms │    no change │
│ Q11   │ 3413.20 ms │      3429.79 ms │    no change │
└───────┴────────────┴─────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary              ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 40063.28ms │
│ Total Time (sort-background)   │ 42044.70ms │
│ Average Time (HEAD)            │  3642.12ms │
│ Average Time (sort-background) │  3822.25ms │
│ Queries Faster                 │          0 │
│ Queries Slower                 │          5 │
│ Queries with No Change         │          6 │
│ Queries with Failure           │          0 │
└────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and sort-background
--------------------
Benchmark sort_tpch10.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃                                   HEAD ┃                        sort-background ┃       Change ┃
┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ Q1    │   1403.13 / 1407.91 ±3.25 / 1412.62 ms │  1697.90 / 1707.41 ±10.72 / 1725.74 ms │ 1.21x slower │
│ Q2    │  1281.30 / 1330.61 ±28.97 / 1369.94 ms │  1670.01 / 1698.72 ±19.26 / 1728.93 ms │ 1.28x slower │
│ Q3    │  8521.15 / 8607.73 ±79.32 / 8711.10 ms │  8471.71 / 8569.07 ±81.34 / 8687.08 ms │    no change │
│ Q4    │  2209.82 / 2227.61 ±19.77 / 2261.03 ms │  2607.15 / 2646.49 ±25.66 / 2685.77 ms │ 1.19x slower │
│ Q5    │  2391.88 / 2409.24 ±18.73 / 2444.09 ms │  2416.08 / 2438.95 ±17.17 / 2460.78 ms │    no change │
│ Q6    │  2544.64 / 2555.34 ±10.27 / 2572.90 ms │  2568.14 / 2602.03 ±18.77 / 2623.91 ms │    no change │
│ Q7    │  4972.19 / 5008.65 ±29.99 / 5048.15 ms │  5050.45 / 5102.12 ±26.88 / 5123.35 ms │    no change │
│ Q8    │  3454.39 / 3559.89 ±72.57 / 3648.60 ms │  3772.64 / 3918.83 ±80.86 / 4010.66 ms │ 1.10x slower │
│ Q9    │  3833.18 / 3862.83 ±24.18 / 3906.60 ms │  4071.69 / 4180.91 ±55.29 / 4224.76 ms │ 1.08x slower │
│ Q10   │ 6038.40 / 6307.66 ±154.80 / 6463.71 ms │ 6289.16 / 6701.86 ±219.39 / 6883.81 ms │ 1.06x slower │
│ Q11   │  3413.20 / 3446.59 ±27.84 / 3497.79 ms │  3429.79 / 3457.36 ±21.92 / 3479.90 ms │    no change │
└───────┴────────────────────────────────────────┴────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary              ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 40724.05ms │
│ Total Time (sort-background)   │ 43023.74ms │
│ Average Time (HEAD)            │  3702.19ms │
│ Average Time (sort-background) │  3911.25ms │
│ Queries Faster                 │          0 │
│ Queries Slower                 │          6 │
│ Queries with No Change         │          5 │
│ Queries with Failure           │          0 │
└────────────────────────────────┴────────────┘

Resource Usage

sort_tpch10 — base (merge-base)

Metric Value
Wall time 205.0s
Peak memory 18.8 GiB
Avg memory 7.1 GiB
CPU user 846.1s
CPU sys 73.3s
Peak spill 0 B

sort_tpch10 — branch

Metric Value
Wall time 220.0s
Peak memory 19.1 GiB
Avg memory 6.8 GiB
CPU user 877.0s
CPU sys 76.4s
Peak spill 0 B

File an issue against this benchmark runner

@rluvaton

rluvaton commented Oct 20, 2026

Copy link
Copy Markdown
Member Author

run benchmarks sort_tpch sort_tpch10 tpch tpch10 tpcds

env:
   prefer_hash_join: false

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6292669244-3641-57gvl 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort_tpch10
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6292669244-3643-tjn45 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpch10
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6292669244-3644-l5fcw 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpcds
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

1 similar comment
@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6292669244-3644-l5fcw 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpcds
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6292669244-3642-sglh7 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpch
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6292669244-3644-l5fcw 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpcds
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6292669244-3640-tnb48 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort_tpch
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpch
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and sort-background
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ sort-background ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 41.46 ms │        40.52 ms │ no change │
│ QQuery 2  │ 18.83 ms │        18.33 ms │ no change │
│ QQuery 3  │ 28.84 ms │        28.70 ms │ no change │
│ QQuery 4  │ 17.57 ms │        17.21 ms │ no change │
│ QQuery 5  │ 35.60 ms │        35.22 ms │ no change │
│ QQuery 6  │ 13.75 ms │        13.52 ms │ no change │
│ QQuery 7  │ 42.75 ms │        41.81 ms │ no change │
│ QQuery 8  │ 41.34 ms │        40.44 ms │ no change │
│ QQuery 9  │ 50.55 ms │        50.79 ms │ no change │
│ QQuery 10 │ 40.91 ms │        40.40 ms │ no change │
│ QQuery 11 │ 13.42 ms │        13.18 ms │ no change │
│ QQuery 12 │ 18.24 ms │        18.24 ms │ no change │
│ QQuery 13 │ 39.85 ms │        38.96 ms │ no change │
│ QQuery 14 │ 24.85 ms │        24.61 ms │ no change │
│ QQuery 15 │ 31.18 ms │        30.83 ms │ no change │
│ QQuery 16 │ 13.97 ms │        13.57 ms │ no change │
│ QQuery 17 │ 75.09 ms │        73.80 ms │ no change │
│ QQuery 18 │ 45.35 ms │        44.73 ms │ no change │
│ QQuery 19 │ 31.75 ms │        31.83 ms │ no change │
│ QQuery 20 │ 31.72 ms │        31.95 ms │ no change │
│ QQuery 21 │ 55.11 ms │        55.34 ms │ no change │
│ QQuery 22 │ 13.67 ms │        13.44 ms │ no change │
└───────────┴──────────┴─────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary              ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)              │ 725.80ms │
│ Total Time (sort-background)   │ 717.41ms │
│ Average Time (HEAD)            │  32.99ms │
│ Average Time (sort-background) │  32.61ms │
│ Queries Faster                 │        0 │
│ Queries Slower                 │        0 │
│ Queries with No Change         │       22 │
│ Queries with Failure           │        0 │
└────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and sort-background
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃                sort-background ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 41.46 / 42.21 ±0.97 / 43.98 ms │ 40.52 / 41.15 ±0.99 / 43.13 ms │ no change │
│ QQuery 2  │ 18.83 / 19.03 ±0.19 / 19.36 ms │ 18.33 / 18.56 ±0.24 / 18.98 ms │ no change │
│ QQuery 3  │ 28.84 / 30.04 ±1.15 / 31.71 ms │ 28.70 / 28.92 ±0.21 / 29.28 ms │ no change │
│ QQuery 4  │ 17.57 / 18.12 ±0.50 / 19.05 ms │ 17.21 / 17.40 ±0.11 / 17.51 ms │ no change │
│ QQuery 5  │ 35.60 / 36.26 ±1.10 / 38.46 ms │ 35.22 / 35.65 ±0.25 / 35.97 ms │ no change │
│ QQuery 6  │ 13.75 / 14.51 ±0.62 / 15.46 ms │ 13.52 / 14.01 ±0.68 / 15.33 ms │ no change │
│ QQuery 7  │ 42.75 / 43.99 ±0.80 / 44.92 ms │ 41.81 / 42.90 ±0.90 / 44.27 ms │ no change │
│ QQuery 8  │ 41.34 / 41.81 ±0.47 / 42.44 ms │ 40.44 / 42.97 ±2.42 / 47.45 ms │ no change │
│ QQuery 9  │ 50.55 / 52.43 ±1.31 / 54.20 ms │ 50.79 / 51.67 ±0.84 / 53.24 ms │ no change │
│ QQuery 10 │ 40.91 / 41.62 ±1.27 / 44.15 ms │ 40.40 / 40.84 ±0.37 / 41.28 ms │ no change │
│ QQuery 11 │ 13.42 / 14.01 ±0.72 / 15.34 ms │ 13.18 / 13.45 ±0.26 / 13.94 ms │ no change │
│ QQuery 12 │ 18.24 / 18.45 ±0.15 / 18.72 ms │ 18.24 / 18.60 ±0.43 / 19.42 ms │ no change │
│ QQuery 13 │ 39.85 / 40.36 ±0.77 / 41.88 ms │ 38.96 / 39.98 ±1.05 / 42.01 ms │ no change │
│ QQuery 14 │ 24.85 / 25.88 ±1.14 / 27.50 ms │ 24.61 / 24.72 ±0.08 / 24.82 ms │ no change │
│ QQuery 15 │ 31.18 / 31.77 ±0.73 / 33.17 ms │ 30.83 / 31.16 ±0.36 / 31.77 ms │ no change │
│ QQuery 16 │ 13.97 / 14.19 ±0.19 / 14.44 ms │ 13.57 / 14.09 ±0.29 / 14.44 ms │ no change │
│ QQuery 17 │ 75.09 / 75.42 ±0.63 / 76.67 ms │ 73.80 / 74.77 ±0.74 / 75.82 ms │ no change │
│ QQuery 18 │ 45.35 / 46.26 ±0.88 / 47.76 ms │ 44.73 / 45.30 ±0.69 / 46.47 ms │ no change │
│ QQuery 19 │ 31.75 / 32.04 ±0.29 / 32.57 ms │ 31.83 / 32.28 ±0.46 / 33.14 ms │ no change │
│ QQuery 20 │ 31.72 / 32.17 ±0.37 / 32.80 ms │ 31.95 / 32.25 ±0.21 / 32.59 ms │ no change │
│ QQuery 21 │ 55.11 / 56.42 ±1.28 / 58.10 ms │ 55.34 / 55.77 ±0.37 / 56.46 ms │ no change │
│ QQuery 22 │ 13.67 / 14.08 ±0.26 / 14.45 ms │ 13.44 / 13.77 ±0.31 / 14.26 ms │ no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary              ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)              │ 741.05ms │
│ Total Time (sort-background)   │ 730.20ms │
│ Average Time (HEAD)            │  33.68ms │
│ Average Time (sort-background) │  33.19ms │
│ Queries Faster                 │        0 │
│ Queries Slower                 │        0 │
│ Queries with No Change         │       22 │
│ Queries with Failure           │        0 │
└────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 483.1 MiB
CPU user 20.4s
CPU sys 1.6s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 475.6 MiB
CPU user 20.0s
CPU sys 1.7s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort_tpch
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and sort-background
--------------------
Benchmark sort_tpch1.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃      HEAD ┃ sort-background ┃       Change ┃
┡━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ Q1    │ 110.88 ms │       114.57 ms │    no change │
│ Q2    │  96.64 ms │        99.30 ms │    no change │
│ Q3    │ 632.68 ms │       657.40 ms │    no change │
│ Q4    │ 160.19 ms │       168.28 ms │ 1.05x slower │
│ Q5    │ 232.66 ms │       254.57 ms │ 1.09x slower │
│ Q6    │ 245.28 ms │       273.90 ms │ 1.12x slower │
│ Q7    │ 413.30 ms │       431.07 ms │    no change │
│ Q8    │ 295.49 ms │       300.59 ms │    no change │
│ Q9    │ 309.62 ms │       318.36 ms │    no change │
│ Q10   │ 454.75 ms │       479.16 ms │ 1.05x slower │
│ Q11   │ 245.47 ms │       248.49 ms │    no change │
└───────┴───────────┴─────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 3196.96ms │
│ Total Time (sort-background)   │ 3345.70ms │
│ Average Time (HEAD)            │  290.63ms │
│ Average Time (sort-background) │  304.15ms │
│ Queries Faster                 │         0 │
│ Queries Slower                 │         4 │
│ Queries with No Change         │         7 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and sort-background
--------------------
Benchmark sort_tpch1.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃                               HEAD ┃                    sort-background ┃       Change ┃
┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ Q1    │  110.88 / 111.96 ±0.91 / 113.46 ms │  114.57 / 115.83 ±1.58 / 118.91 ms │    no change │
│ Q2    │    96.64 / 97.96 ±1.46 / 100.81 ms │   99.30 / 101.29 ±1.88 / 104.86 ms │    no change │
│ Q3    │  632.68 / 640.62 ±5.14 / 647.74 ms │  657.40 / 667.93 ±6.03 / 675.99 ms │    no change │
│ Q4    │  160.19 / 161.38 ±1.04 / 162.76 ms │  168.28 / 169.90 ±1.20 / 171.59 ms │ 1.05x slower │
│ Q5    │  232.66 / 234.17 ±1.02 / 235.49 ms │  254.57 / 259.01 ±4.22 / 264.23 ms │ 1.11x slower │
│ Q6    │  245.28 / 246.42 ±0.82 / 247.59 ms │  273.90 / 278.52 ±5.54 / 289.15 ms │ 1.13x slower │
│ Q7    │ 413.30 / 423.91 ±14.53 / 452.33 ms │ 431.07 / 438.33 ±11.41 / 460.96 ms │    no change │
│ Q8    │ 295.49 / 304.69 ±12.95 / 329.11 ms │  300.59 / 315.69 ±9.95 / 327.29 ms │    no change │
│ Q9    │  309.62 / 320.17 ±8.20 / 334.90 ms │ 318.36 / 333.25 ±11.01 / 343.42 ms │    no change │
│ Q10   │ 454.75 / 469.82 ±16.51 / 501.86 ms │ 479.16 / 492.98 ±13.01 / 510.49 ms │    no change │
│ Q11   │ 245.47 / 255.43 ±10.63 / 275.28 ms │  248.49 / 258.70 ±9.95 / 271.30 ms │    no change │
└───────┴────────────────────────────────────┴────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 3266.53ms │
│ Total Time (sort-background)   │ 3431.43ms │
│ Average Time (HEAD)            │  296.96ms │
│ Average Time (sort-background) │  311.95ms │
│ Queries Faster                 │         0 │
│ Queries Slower                 │         3 │
│ Queries with No Change         │         8 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Resource Usage

sort_tpch — base (merge-base)

Metric Value
Wall time 20.0s
Peak memory 2.5 GiB
Avg memory 1.1 GiB
CPU user 60.9s
CPU sys 3.0s
Peak spill 0 B

sort_tpch — branch

Metric Value
Wall time 20.0s
Peak memory 2.4 GiB
Avg memory 1.1 GiB
CPU user 63.6s
CPU sys 3.6s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpch10
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and sort-background
--------------------
Benchmark tpch_sf10.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃      HEAD ┃ sort-background ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 332.10 ms │       337.74 ms │ no change │
│ QQuery 2  │  93.18 ms │        92.27 ms │ no change │
│ QQuery 3  │ 224.58 ms │       225.97 ms │ no change │
│ QQuery 4  │ 116.68 ms │       117.99 ms │ no change │
│ QQuery 5  │ 369.78 ms │       364.02 ms │ no change │
│ QQuery 6  │ 103.41 ms │       103.09 ms │ no change │
│ QQuery 7  │ 464.61 ms │       465.54 ms │ no change │
│ QQuery 8  │ 367.48 ms │       362.44 ms │ no change │
│ QQuery 9  │ 548.70 ms │       538.88 ms │ no change │
│ QQuery 10 │ 304.33 ms │       295.99 ms │ no change │
│ QQuery 11 │  66.55 ms │        65.34 ms │ no change │
│ QQuery 12 │ 121.66 ms │       118.08 ms │ no change │
│ QQuery 13 │ 327.83 ms │       318.03 ms │ no change │
│ QQuery 14 │ 172.86 ms │       174.95 ms │ no change │
│ QQuery 15 │ 306.75 ms │       303.53 ms │ no change │
│ QQuery 16 │  65.93 ms │        65.32 ms │ no change │
│ QQuery 17 │ 613.93 ms │       624.59 ms │ no change │
│ QQuery 18 │ 763.91 ms │       766.17 ms │ no change │
│ QQuery 19 │ 240.43 ms │       235.32 ms │ no change │
│ QQuery 20 │ 272.38 ms │       267.87 ms │ no change │
│ QQuery 21 │ 688.01 ms │       682.08 ms │ no change │
│ QQuery 22 │  63.91 ms │        62.18 ms │ no change │
└───────────┴───────────┴─────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 6629.00ms │
│ Total Time (sort-background)   │ 6587.40ms │
│ Average Time (HEAD)            │  301.32ms │
│ Average Time (sort-background) │  299.43ms │
│ Queries Faster                 │         0 │
│ Queries Slower                 │         0 │
│ Queries with No Change         │        22 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and sort-background
--------------------
Benchmark tpch_sf10.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                               HEAD ┃                    sort-background ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │  332.10 / 334.65 ±1.60 / 336.40 ms │  337.74 / 338.56 ±0.75 / 339.81 ms │ no change │
│ QQuery 2  │     93.18 / 95.08 ±1.63 / 97.59 ms │    92.27 / 96.13 ±3.01 / 100.56 ms │ no change │
│ QQuery 3  │  224.58 / 228.49 ±2.94 / 232.96 ms │  225.97 / 231.19 ±3.50 / 236.86 ms │ no change │
│ QQuery 4  │  116.68 / 118.64 ±1.29 / 119.90 ms │  117.99 / 119.80 ±1.41 / 122.12 ms │ no change │
│ QQuery 5  │  369.78 / 373.56 ±3.92 / 381.08 ms │  364.02 / 372.48 ±8.33 / 384.31 ms │ no change │
│ QQuery 6  │  103.41 / 105.51 ±1.67 / 108.49 ms │  103.09 / 105.11 ±2.26 / 109.40 ms │ no change │
│ QQuery 7  │  464.61 / 472.03 ±5.70 / 482.08 ms │  465.54 / 480.65 ±9.10 / 492.05 ms │ no change │
│ QQuery 8  │  367.48 / 370.93 ±2.46 / 374.96 ms │  362.44 / 367.00 ±3.15 / 372.06 ms │ no change │
│ QQuery 9  │ 548.70 / 563.74 ±16.50 / 594.44 ms │ 538.88 / 558.35 ±16.03 / 575.97 ms │ no change │
│ QQuery 10 │ 304.33 / 324.59 ±16.48 / 350.42 ms │  295.99 / 309.14 ±9.59 / 324.42 ms │ no change │
│ QQuery 11 │     66.55 / 68.79 ±1.98 / 71.97 ms │     65.34 / 70.05 ±7.20 / 84.33 ms │ no change │
│ QQuery 12 │  121.66 / 128.53 ±9.04 / 146.14 ms │  118.08 / 123.39 ±7.30 / 137.76 ms │ no change │
│ QQuery 13 │  327.83 / 334.55 ±7.93 / 348.84 ms │  318.03 / 323.53 ±4.35 / 329.83 ms │ no change │
│ QQuery 14 │  172.86 / 180.19 ±5.91 / 188.58 ms │  174.95 / 180.32 ±5.57 / 187.62 ms │ no change │
│ QQuery 15 │  306.75 / 309.08 ±3.13 / 315.22 ms │  303.53 / 308.81 ±3.68 / 313.06 ms │ no change │
│ QQuery 16 │     65.93 / 67.35 ±1.37 / 69.86 ms │     65.32 / 70.42 ±5.86 / 81.56 ms │ no change │
│ QQuery 17 │  613.93 / 624.45 ±5.41 / 628.59 ms │  624.59 / 632.95 ±9.36 / 647.71 ms │ no change │
│ QQuery 18 │ 763.91 / 781.02 ±16.51 / 806.32 ms │ 766.17 / 782.30 ±13.32 / 806.27 ms │ no change │
│ QQuery 19 │  240.43 / 256.01 ±8.78 / 266.80 ms │ 235.32 / 249.40 ±13.46 / 271.96 ms │ no change │
│ QQuery 20 │  272.38 / 280.40 ±4.84 / 286.88 ms │  267.87 / 276.92 ±6.63 / 288.03 ms │ no change │
│ QQuery 21 │  688.01 / 693.49 ±4.95 / 701.23 ms │  682.08 / 688.62 ±4.33 / 693.79 ms │ no change │
│ QQuery 22 │     63.91 / 67.75 ±2.21 / 70.64 ms │     62.18 / 66.92 ±2.62 / 70.14 ms │ no change │
└───────────┴────────────────────────────────────┴────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 6778.83ms │
│ Total Time (sort-background)   │ 6752.04ms │
│ Average Time (HEAD)            │  308.13ms │
│ Average Time (sort-background) │  306.91ms │
│ Queries Faster                 │         0 │
│ Queries Slower                 │         0 │
│ Queries with No Change         │        22 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Resource Usage

tpch10 — base (merge-base)

Metric Value
Wall time 35.0s
Peak memory 5.0 GiB
Avg memory 1.5 GiB
CPU user 344.8s
CPU sys 19.6s
Peak spill 0 B

tpch10 — branch

Metric Value
Wall time 35.0s
Peak memory 4.9 GiB
Avg memory 1.5 GiB
CPU user 343.1s
CPU sys 19.7s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark tpcds
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and sort-background
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ sort-background ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │    5.12 ms │         5.28 ms │ no change │
│ QQuery 2  │   80.90 ms │        81.14 ms │ no change │
│ QQuery 3  │   28.63 ms │        28.76 ms │ no change │
│ QQuery 4  │  450.68 ms │       457.92 ms │ no change │
│ QQuery 5  │   48.62 ms │        49.23 ms │ no change │
│ QQuery 6  │   34.15 ms │        35.16 ms │ no change │
│ QQuery 7  │   71.53 ms │        72.22 ms │ no change │
│ QQuery 8  │   36.02 ms │        35.68 ms │ no change │
│ QQuery 9  │   52.93 ms │        51.79 ms │ no change │
│ QQuery 10 │   59.65 ms │        60.13 ms │ no change │
│ QQuery 11 │  277.70 ms │       283.15 ms │ no change │
│ QQuery 12 │   28.19 ms │        28.27 ms │ no change │
│ QQuery 13 │  112.21 ms │       113.44 ms │ no change │
│ QQuery 14 │  380.80 ms │       382.59 ms │ no change │
│ QQuery 15 │   53.74 ms │        53.94 ms │ no change │
│ QQuery 16 │    5.79 ms │         6.06 ms │ no change │
│ QQuery 17 │   77.49 ms │        77.69 ms │ no change │
│ QQuery 18 │  102.35 ms │       103.95 ms │ no change │
│ QQuery 19 │   40.42 ms │        40.95 ms │ no change │
│ QQuery 20 │   34.68 ms │        35.19 ms │ no change │
│ QQuery 21 │   17.08 ms │        17.00 ms │ no change │
│ QQuery 22 │   63.28 ms │        65.13 ms │ no change │
│ QQuery 23 │  300.78 ms │       306.51 ms │ no change │
│ QQuery 24 │  193.64 ms │       193.63 ms │ no change │
│ QQuery 25 │  105.84 ms │       106.34 ms │ no change │
│ QQuery 26 │   45.82 ms │        45.30 ms │ no change │
│ QQuery 27 │    5.70 ms │         5.85 ms │ no change │
│ QQuery 28 │   56.15 ms │        57.31 ms │ no change │
│ QQuery 29 │   94.36 ms │        94.88 ms │ no change │
│ QQuery 30 │   32.34 ms │        32.23 ms │ no change │
│ QQuery 31 │  107.26 ms │       107.11 ms │ no change │
│ QQuery 32 │   19.76 ms │        20.21 ms │ no change │
│ QQuery 33 │   35.88 ms │        36.31 ms │ no change │
│ QQuery 34 │   10.18 ms │        10.13 ms │ no change │
│ QQuery 35 │   71.79 ms │        71.93 ms │ no change │
│ QQuery 36 │    5.54 ms │         5.61 ms │ no change │
│ QQuery 37 │    6.68 ms │         6.66 ms │ no change │
│ QQuery 38 │   60.77 ms │        60.84 ms │ no change │
│ QQuery 39 │   72.69 ms │        72.14 ms │ no change │
│ QQuery 40 │   23.49 ms │        23.53 ms │ no change │
│ QQuery 41 │   12.32 ms │        12.27 ms │ no change │
│ QQuery 42 │   23.55 ms │        23.64 ms │ no change │
│ QQuery 43 │    4.74 ms │         4.64 ms │ no change │
│ QQuery 44 │    8.55 ms │         8.42 ms │ no change │
│ QQuery 45 │   35.85 ms │        36.42 ms │ no change │
│ QQuery 46 │   11.69 ms │        11.42 ms │ no change │
│ QQuery 47 │  222.71 ms │       225.87 ms │ no change │
│ QQuery 48 │   90.02 ms │        91.21 ms │ no change │
│ QQuery 49 │   69.54 ms │        68.96 ms │ no change │
│ QQuery 50 │   57.83 ms │        57.43 ms │ no change │
│ QQuery 51 │   91.39 ms │        87.05 ms │ no change │
│ QQuery 52 │   24.10 ms │        23.68 ms │ no change │
│ QQuery 53 │   28.99 ms │        28.84 ms │ no change │
│ QQuery 54 │   52.86 ms │        53.41 ms │ no change │
│ QQuery 55 │   23.72 ms │        23.60 ms │ no change │
│ QQuery 56 │   37.72 ms │        37.32 ms │ no change │
│ QQuery 57 │  163.87 ms │       164.75 ms │ no change │
│ QQuery 58 │  106.94 ms │       106.78 ms │ no change │
│ QQuery 59 │  115.09 ms │       114.85 ms │ no change │
│ QQuery 60 │   37.79 ms │        38.08 ms │ no change │
│ QQuery 61 │   10.40 ms │        10.41 ms │ no change │
│ QQuery 62 │   43.77 ms │        43.80 ms │ no change │
│ QQuery 63 │   29.34 ms │        29.20 ms │ no change │
│ QQuery 64 │  351.10 ms │       354.47 ms │ no change │
│ QQuery 65 │  128.29 ms │       126.93 ms │ no change │
│ QQuery 66 │   74.35 ms │        74.24 ms │ no change │
│ QQuery 67 │  251.27 ms │       255.12 ms │ no change │
│ QQuery 68 │   11.93 ms │        12.24 ms │ no change │
│ QQuery 69 │   55.47 ms │        55.25 ms │ no change │
│ QQuery 70 │  103.50 ms │       102.20 ms │ no change │
│ QQuery 71 │   34.68 ms │        34.58 ms │ no change │
│ QQuery 72 │ 1650.04 ms │      1649.39 ms │ no change │
│ QQuery 73 │   10.06 ms │         9.95 ms │ no change │
│ QQuery 74 │  160.68 ms │       163.89 ms │ no change │
│ QQuery 75 │  139.11 ms │       138.60 ms │ no change │
│ QQuery 76 │   34.18 ms │        33.89 ms │ no change │
│ QQuery 77 │   59.42 ms │        59.46 ms │ no change │
│ QQuery 78 │  161.78 ms │       160.41 ms │ no change │
│ QQuery 79 │   65.37 ms │        66.11 ms │ no change │
│ QQuery 80 │   93.36 ms │        93.89 ms │ no change │
│ QQuery 81 │   25.64 ms │        25.81 ms │ no change │
│ QQuery 82 │   16.54 ms │        16.52 ms │ no change │
│ QQuery 83 │   32.64 ms │        33.17 ms │ no change │
│ QQuery 84 │   29.58 ms │        29.56 ms │ no change │
│ QQuery 85 │   98.35 ms │        98.75 ms │ no change │
│ QQuery 86 │   25.71 ms │        26.71 ms │ no change │
│ QQuery 87 │   62.20 ms │        61.72 ms │ no change │
│ QQuery 88 │   58.60 ms │        58.60 ms │ no change │
│ QQuery 89 │   34.87 ms │        35.59 ms │ no change │
│ QQuery 90 │   16.29 ms │        16.46 ms │ no change │
│ QQuery 91 │   40.60 ms │        41.00 ms │ no change │
│ QQuery 92 │   29.05 ms │        30.08 ms │ no change │
│ QQuery 93 │   50.00 ms │        48.73 ms │ no change │
│ QQuery 94 │   38.53 ms │        38.85 ms │ no change │
│ QQuery 95 │   80.74 ms │        80.55 ms │ no change │
│ QQuery 96 │   23.70 ms │        23.77 ms │ no change │
│ QQuery 97 │   50.16 ms │        50.63 ms │ no change │
│ QQuery 98 │   43.40 ms │        43.51 ms │ no change │
│ QQuery 99 │   64.75 ms │        65.30 ms │ no change │
└───────────┴────────────┴─────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 8819.41ms │
│ Total Time (sort-background)   │ 8859.17ms │
│ Average Time (HEAD)            │   89.08ms │
│ Average Time (sort-background) │   89.49ms │
│ Queries Faster                 │         0 │
│ Queries Slower                 │         0 │
│ Queries with No Change         │        99 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and sort-background
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                       sort-background ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.12 / 5.74 ±1.08 / 7.89 ms │           5.28 / 5.84 ±0.96 / 7.75 ms │     no change │
│ QQuery 2  │        80.90 / 81.20 ±0.19 / 81.43 ms │        81.14 / 81.41 ±0.24 / 81.75 ms │     no change │
│ QQuery 3  │        28.63 / 28.89 ±0.23 / 29.31 ms │        28.76 / 28.97 ±0.17 / 29.12 ms │     no change │
│ QQuery 4  │     450.68 / 453.78 ±3.56 / 460.41 ms │     457.92 / 467.04 ±8.56 / 482.50 ms │     no change │
│ QQuery 5  │        48.62 / 51.47 ±3.48 / 58.29 ms │        49.23 / 49.96 ±0.79 / 51.45 ms │     no change │
│ QQuery 6  │        34.15 / 34.96 ±0.65 / 35.78 ms │        35.16 / 35.40 ±0.24 / 35.78 ms │     no change │
│ QQuery 7  │        71.53 / 72.27 ±0.60 / 73.08 ms │        72.22 / 72.74 ±0.45 / 73.53 ms │     no change │
│ QQuery 8  │        36.02 / 36.16 ±0.20 / 36.55 ms │        35.68 / 36.13 ±0.33 / 36.71 ms │     no change │
│ QQuery 9  │        52.93 / 54.08 ±0.86 / 54.85 ms │        51.79 / 55.18 ±4.20 / 63.34 ms │     no change │
│ QQuery 10 │        59.65 / 60.52 ±0.59 / 61.26 ms │        60.13 / 60.59 ±0.45 / 61.40 ms │     no change │
│ QQuery 11 │     277.70 / 279.54 ±1.42 / 282.09 ms │     283.15 / 287.03 ±2.32 / 289.24 ms │     no change │
│ QQuery 12 │        28.19 / 29.07 ±0.55 / 29.74 ms │        28.27 / 30.05 ±1.91 / 33.76 ms │     no change │
│ QQuery 13 │     112.21 / 114.28 ±1.41 / 116.21 ms │     113.44 / 114.87 ±1.03 / 116.45 ms │     no change │
│ QQuery 14 │     380.80 / 385.95 ±3.32 / 390.00 ms │     382.59 / 386.07 ±2.88 / 391.15 ms │     no change │
│ QQuery 15 │        53.74 / 54.32 ±0.50 / 55.13 ms │        53.94 / 54.82 ±1.06 / 56.80 ms │     no change │
│ QQuery 16 │           5.79 / 5.97 ±0.32 / 6.60 ms │           6.06 / 6.21 ±0.21 / 6.63 ms │     no change │
│ QQuery 17 │        77.49 / 78.85 ±1.27 / 81.04 ms │        77.69 / 79.19 ±1.53 / 82.12 ms │     no change │
│ QQuery 18 │     102.35 / 103.77 ±0.88 / 104.70 ms │     103.95 / 104.91 ±0.82 / 105.97 ms │     no change │
│ QQuery 19 │        40.42 / 41.11 ±0.65 / 41.90 ms │        40.95 / 41.69 ±0.55 / 42.61 ms │     no change │
│ QQuery 20 │        34.68 / 36.03 ±1.86 / 39.68 ms │        35.19 / 36.71 ±1.85 / 40.36 ms │     no change │
│ QQuery 21 │        17.08 / 17.20 ±0.08 / 17.31 ms │        17.00 / 17.27 ±0.17 / 17.47 ms │     no change │
│ QQuery 22 │        63.28 / 63.96 ±0.53 / 64.73 ms │        65.13 / 67.84 ±1.56 / 69.64 ms │  1.06x slower │
│ QQuery 23 │     300.78 / 305.69 ±4.78 / 313.22 ms │     306.51 / 314.46 ±6.49 / 322.60 ms │     no change │
│ QQuery 24 │     193.64 / 196.89 ±5.13 / 207.00 ms │     193.63 / 199.94 ±7.80 / 215.27 ms │     no change │
│ QQuery 25 │     105.84 / 106.57 ±0.58 / 107.62 ms │     106.34 / 108.55 ±1.89 / 111.01 ms │     no change │
│ QQuery 26 │        45.82 / 47.37 ±2.61 / 52.58 ms │        45.30 / 45.49 ±0.16 / 45.69 ms │     no change │
│ QQuery 27 │           5.70 / 5.87 ±0.19 / 6.22 ms │           5.85 / 5.98 ±0.13 / 6.23 ms │     no change │
│ QQuery 28 │        56.15 / 59.24 ±2.44 / 61.96 ms │        57.31 / 61.09 ±2.69 / 65.74 ms │     no change │
│ QQuery 29 │       94.36 / 96.74 ±2.46 / 100.37 ms │       94.88 / 98.52 ±4.57 / 107.24 ms │     no change │
│ QQuery 30 │        32.34 / 32.75 ±0.23 / 33.06 ms │        32.23 / 32.63 ±0.44 / 33.37 ms │     no change │
│ QQuery 31 │     107.26 / 107.67 ±0.51 / 108.66 ms │     107.11 / 108.42 ±1.12 / 110.33 ms │     no change │
│ QQuery 32 │        19.76 / 20.07 ±0.19 / 20.34 ms │        20.21 / 20.32 ±0.07 / 20.44 ms │     no change │
│ QQuery 33 │        35.88 / 37.19 ±0.87 / 38.23 ms │        36.31 / 37.03 ±0.69 / 38.10 ms │     no change │
│ QQuery 34 │        10.18 / 12.47 ±4.07 / 20.61 ms │        10.13 / 10.75 ±0.65 / 11.88 ms │ +1.16x faster │
│ QQuery 35 │        71.79 / 72.32 ±0.47 / 73.07 ms │        71.93 / 73.67 ±2.39 / 78.34 ms │     no change │
│ QQuery 36 │           5.54 / 5.73 ±0.23 / 6.18 ms │           5.61 / 5.77 ±0.15 / 6.05 ms │     no change │
│ QQuery 37 │           6.68 / 6.74 ±0.05 / 6.81 ms │           6.66 / 6.75 ±0.09 / 6.90 ms │     no change │
│ QQuery 38 │        60.77 / 61.82 ±0.70 / 62.47 ms │        60.84 / 61.27 ±0.33 / 61.75 ms │     no change │
│ QQuery 39 │        72.69 / 75.23 ±3.64 / 82.39 ms │        72.14 / 73.50 ±1.17 / 75.60 ms │     no change │
│ QQuery 40 │        23.49 / 24.30 ±0.51 / 24.84 ms │        23.53 / 24.26 ±0.58 / 25.19 ms │     no change │
│ QQuery 41 │        12.32 / 13.39 ±0.74 / 14.63 ms │        12.27 / 12.98 ±1.08 / 15.12 ms │     no change │
│ QQuery 42 │        23.55 / 23.95 ±0.31 / 24.36 ms │        23.64 / 24.22 ±0.55 / 25.25 ms │     no change │
│ QQuery 43 │           4.74 / 5.36 ±0.89 / 7.11 ms │           4.64 / 4.79 ±0.25 / 5.28 ms │ +1.12x faster │
│ QQuery 44 │           8.55 / 8.66 ±0.07 / 8.75 ms │           8.42 / 8.60 ±0.10 / 8.70 ms │     no change │
│ QQuery 45 │        35.85 / 37.42 ±0.99 / 38.37 ms │        36.42 / 37.32 ±0.57 / 38.06 ms │     no change │
│ QQuery 46 │        11.69 / 11.98 ±0.26 / 12.47 ms │        11.42 / 11.56 ±0.09 / 11.65 ms │     no change │
│ QQuery 47 │     222.71 / 230.42 ±7.63 / 240.57 ms │     225.87 / 231.33 ±4.05 / 236.96 ms │     no change │
│ QQuery 48 │        90.02 / 92.91 ±3.11 / 98.23 ms │        91.21 / 92.70 ±1.35 / 95.25 ms │     no change │
│ QQuery 49 │        69.54 / 70.64 ±1.18 / 72.85 ms │        68.96 / 69.71 ±0.78 / 71.11 ms │     no change │
│ QQuery 50 │        57.83 / 60.08 ±3.85 / 67.74 ms │        57.43 / 59.78 ±3.38 / 66.49 ms │     no change │
│ QQuery 51 │       91.39 / 94.17 ±3.82 / 101.71 ms │        87.05 / 88.85 ±1.54 / 91.71 ms │ +1.06x faster │
│ QQuery 52 │        24.10 / 24.34 ±0.24 / 24.75 ms │        23.68 / 24.08 ±0.43 / 24.91 ms │     no change │
│ QQuery 53 │        28.99 / 29.26 ±0.19 / 29.48 ms │        28.84 / 29.03 ±0.15 / 29.29 ms │     no change │
│ QQuery 54 │        52.86 / 55.11 ±3.92 / 62.94 ms │        53.41 / 53.91 ±0.47 / 54.48 ms │     no change │
│ QQuery 55 │        23.72 / 24.06 ±0.32 / 24.63 ms │        23.60 / 23.88 ±0.36 / 24.57 ms │     no change │
│ QQuery 56 │        37.72 / 38.19 ±0.41 / 38.74 ms │        37.32 / 38.54 ±1.63 / 41.73 ms │     no change │
│ QQuery 57 │     163.87 / 168.64 ±3.82 / 173.57 ms │     164.75 / 170.59 ±4.02 / 175.25 ms │     no change │
│ QQuery 58 │     106.94 / 108.95 ±2.03 / 112.50 ms │     106.78 / 108.12 ±1.72 / 111.47 ms │     no change │
│ QQuery 59 │     115.09 / 117.07 ±2.66 / 122.22 ms │     114.85 / 116.01 ±0.86 / 117.22 ms │     no change │
│ QQuery 60 │        37.79 / 38.29 ±0.42 / 39.06 ms │        38.08 / 38.68 ±0.58 / 39.74 ms │     no change │
│ QQuery 61 │        10.40 / 10.58 ±0.12 / 10.75 ms │        10.41 / 10.67 ±0.16 / 10.80 ms │     no change │
│ QQuery 62 │        43.77 / 44.14 ±0.31 / 44.60 ms │        43.80 / 44.62 ±0.48 / 45.27 ms │     no change │
│ QQuery 63 │        29.34 / 29.77 ±0.23 / 30.01 ms │        29.20 / 30.11 ±1.25 / 32.55 ms │     no change │
│ QQuery 64 │     351.10 / 357.19 ±7.11 / 368.06 ms │     354.47 / 360.02 ±3.62 / 363.38 ms │     no change │
│ QQuery 65 │     128.29 / 130.08 ±1.36 / 131.90 ms │     126.93 / 131.47 ±4.44 / 139.39 ms │     no change │
│ QQuery 66 │        74.35 / 76.61 ±1.84 / 79.15 ms │        74.24 / 75.05 ±0.64 / 75.82 ms │     no change │
│ QQuery 67 │    251.27 / 262.23 ±11.74 / 279.22 ms │     255.12 / 261.89 ±4.48 / 267.78 ms │     no change │
│ QQuery 68 │        11.93 / 15.87 ±7.04 / 29.94 ms │        12.24 / 12.42 ±0.17 / 12.68 ms │ +1.28x faster │
│ QQuery 69 │        55.47 / 56.03 ±0.53 / 56.78 ms │        55.25 / 56.27 ±0.75 / 57.47 ms │     no change │
│ QQuery 70 │     103.50 / 104.96 ±1.07 / 106.78 ms │     102.20 / 106.29 ±6.57 / 119.40 ms │     no change │
│ QQuery 71 │        34.68 / 38.18 ±6.21 / 50.55 ms │        34.58 / 35.68 ±1.18 / 37.97 ms │ +1.07x faster │
│ QQuery 72 │ 1650.04 / 1726.78 ±67.57 / 1841.73 ms │ 1649.39 / 1686.66 ±40.50 / 1751.34 ms │     no change │
│ QQuery 73 │        10.06 / 10.43 ±0.25 / 10.72 ms │         9.95 / 10.37 ±0.39 / 11.05 ms │     no change │
│ QQuery 74 │     160.68 / 165.21 ±4.58 / 173.38 ms │     163.89 / 166.96 ±2.54 / 171.28 ms │     no change │
│ QQuery 75 │     139.11 / 139.63 ±0.48 / 140.30 ms │     138.60 / 141.44 ±3.77 / 148.63 ms │     no change │
│ QQuery 76 │        34.18 / 34.95 ±1.23 / 37.39 ms │        33.89 / 34.47 ±0.31 / 34.71 ms │     no change │
│ QQuery 77 │        59.42 / 60.03 ±0.45 / 60.67 ms │        59.46 / 60.84 ±1.72 / 64.15 ms │     no change │
│ QQuery 78 │     161.78 / 163.62 ±2.95 / 169.50 ms │     160.41 / 164.63 ±2.40 / 167.17 ms │     no change │
│ QQuery 79 │        65.37 / 66.15 ±0.42 / 66.61 ms │        66.11 / 66.70 ±0.76 / 68.16 ms │     no change │
│ QQuery 80 │        93.36 / 95.78 ±1.98 / 99.20 ms │       93.89 / 97.92 ±4.66 / 106.49 ms │     no change │
│ QQuery 81 │        25.64 / 26.26 ±0.76 / 27.74 ms │        25.81 / 26.57 ±1.17 / 28.89 ms │     no change │
│ QQuery 82 │        16.54 / 17.16 ±1.10 / 19.35 ms │        16.52 / 17.18 ±0.57 / 18.18 ms │     no change │
│ QQuery 83 │        32.64 / 32.74 ±0.11 / 32.92 ms │        33.17 / 33.36 ±0.26 / 33.85 ms │     no change │
│ QQuery 84 │        29.58 / 29.95 ±0.42 / 30.53 ms │        29.56 / 29.99 ±0.39 / 30.48 ms │     no change │
│ QQuery 85 │      98.35 / 101.79 ±5.77 / 113.31 ms │      98.75 / 102.18 ±3.98 / 109.76 ms │     no change │
│ QQuery 86 │        25.71 / 26.59 ±1.01 / 28.53 ms │        26.71 / 27.55 ±1.24 / 30.00 ms │     no change │
│ QQuery 87 │        62.20 / 62.71 ±0.41 / 63.20 ms │        61.72 / 62.80 ±0.88 / 64.08 ms │     no change │
│ QQuery 88 │        58.60 / 58.86 ±0.22 / 59.18 ms │        58.60 / 61.21 ±4.81 / 70.81 ms │     no change │
│ QQuery 89 │        34.87 / 35.23 ±0.28 / 35.63 ms │        35.59 / 35.86 ±0.25 / 36.29 ms │     no change │
│ QQuery 90 │        16.29 / 16.52 ±0.14 / 16.66 ms │        16.46 / 16.56 ±0.06 / 16.61 ms │     no change │
│ QQuery 91 │        40.60 / 41.61 ±1.16 / 43.86 ms │        41.00 / 41.82 ±0.58 / 42.73 ms │     no change │
│ QQuery 92 │        29.05 / 29.54 ±0.28 / 29.87 ms │        30.08 / 30.34 ±0.28 / 30.78 ms │     no change │
│ QQuery 93 │        50.00 / 50.80 ±0.58 / 51.55 ms │        48.73 / 49.59 ±0.45 / 50.03 ms │     no change │
│ QQuery 94 │        38.53 / 39.96 ±1.96 / 43.62 ms │        38.85 / 40.53 ±1.59 / 43.42 ms │     no change │
│ QQuery 95 │        80.74 / 81.85 ±0.93 / 83.17 ms │        80.55 / 81.92 ±1.24 / 84.13 ms │     no change │
│ QQuery 96 │        23.70 / 23.87 ±0.14 / 24.04 ms │        23.77 / 23.92 ±0.07 / 23.99 ms │     no change │
│ QQuery 97 │        50.16 / 51.52 ±1.60 / 54.58 ms │        50.63 / 51.74 ±0.91 / 52.64 ms │     no change │
│ QQuery 98 │        43.40 / 44.75 ±1.77 / 48.25 ms │        43.51 / 44.92 ±0.75 / 45.68 ms │     no change │
│ QQuery 99 │        64.75 / 65.55 ±0.64 / 66.64 ms │        65.30 / 66.30 ±1.37 / 69.02 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 9042.13ms │
│ Total Time (sort-background)   │ 9051.84ms │
│ Average Time (HEAD)            │   91.33ms │
│ Average Time (sort-background) │   91.43ms │
│ Queries Faster                 │         5 │
│ Queries Slower                 │         1 │
│ Queries with No Change         │        93 │
│ Queries with Failure           │         0 │
└────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 50.0s
Peak memory 1.8 GiB
Avg memory 1.3 GiB
CPU user 194.8s
CPU sys 5.6s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 50.0s
Peak memory 1.9 GiB
Avg memory 1.3 GiB
CPU user 194.0s
CPU sys 5.8s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort_tpch10
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and sort-background
--------------------
Benchmark sort_tpch10.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃       HEAD ┃ sort-background ┃       Change ┃
┡━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ Q1    │ 1393.25 ms │      1446.49 ms │    no change │
│ Q2    │ 1279.27 ms │      1400.75 ms │ 1.09x slower │
│ Q3    │ 8419.79 ms │      8441.55 ms │    no change │
│ Q4    │ 2190.92 ms │      2224.23 ms │    no change │
│ Q5    │ 2390.70 ms │      2624.78 ms │ 1.10x slower │
│ Q6    │ 2498.80 ms │      2871.47 ms │ 1.15x slower │
│ Q7    │ 4902.93 ms │      4877.07 ms │    no change │
│ Q8    │ 3507.57 ms │      3506.75 ms │    no change │
│ Q9    │ 3843.81 ms │      3807.96 ms │    no change │
│ Q10   │ 5909.93 ms │      6106.64 ms │    no change │
│ Q11   │ 3398.78 ms │      3369.38 ms │    no change │
└───────┴────────────┴─────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary              ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 39735.76ms │
│ Total Time (sort-background)   │ 40677.06ms │
│ Average Time (HEAD)            │  3612.34ms │
│ Average Time (sort-background) │  3697.91ms │
│ Queries Faster                 │          0 │
│ Queries Slower                 │          3 │
│ Queries with No Change         │          8 │
│ Queries with Failure           │          0 │
└────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and sort-background
--------------------
Benchmark sort_tpch10.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃                                   HEAD ┃                        sort-background ┃       Change ┃
┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ Q1    │   1393.25 / 1399.35 ±5.03 / 1407.39 ms │  1446.49 / 1506.08 ±49.80 / 1576.53 ms │ 1.08x slower │
│ Q2    │  1279.27 / 1321.04 ±25.49 / 1358.64 ms │  1400.75 / 1413.04 ±11.13 / 1429.19 ms │ 1.07x slower │
│ Q3    │ 8419.79 / 8539.94 ±121.12 / 8771.12 ms │  8441.55 / 8464.67 ±22.26 / 8502.51 ms │    no change │
│ Q4    │  2190.92 / 2225.17 ±22.87 / 2261.88 ms │  2224.23 / 2280.77 ±35.38 / 2326.25 ms │    no change │
│ Q5    │  2390.70 / 2411.56 ±13.92 / 2427.93 ms │ 2624.78 / 2720.08 ±107.85 / 2920.66 ms │ 1.13x slower │
│ Q6    │  2498.80 / 2528.61 ±15.78 / 2542.46 ms │  2871.47 / 2896.32 ±19.06 / 2926.32 ms │ 1.15x slower │
│ Q7    │  4902.93 / 4928.74 ±22.99 / 4963.78 ms │  4877.07 / 4949.08 ±56.64 / 5034.85 ms │    no change │
│ Q8    │  3507.57 / 3558.54 ±52.44 / 3637.39 ms │  3506.75 / 3571.91 ±46.27 / 3632.32 ms │    no change │
│ Q9    │  3843.81 / 3892.45 ±50.61 / 3983.42 ms │  3807.96 / 3838.39 ±24.39 / 3879.94 ms │    no change │
│ Q10   │ 5909.93 / 6289.28 ±199.28 / 6449.64 ms │ 6106.64 / 6340.85 ±119.95 / 6431.03 ms │    no change │
│ Q11   │  3398.78 / 3431.30 ±31.47 / 3484.28 ms │  3369.38 / 3399.83 ±33.62 / 3464.42 ms │    no change │
└───────┴────────────────────────────────────────┴────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary              ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)              │ 40525.99ms │
│ Total Time (sort-background)   │ 41381.02ms │
│ Average Time (HEAD)            │  3684.18ms │
│ Average Time (sort-background) │  3761.91ms │
│ Queries Faster                 │          0 │
│ Queries Slower                 │          4 │
│ Queries with No Change         │          7 │
│ Queries with Failure           │          0 │
└────────────────────────────────┴────────────┘

Resource Usage

sort_tpch10 — base (merge-base)

Metric Value
Wall time 205.0s
Peak memory 18.9 GiB
Avg memory 7.1 GiB
CPU user 833.9s
CPU sys 72.9s
Peak spill 0 B

sort_tpch10 — branch

Metric Value
Wall time 210.0s
Peak memory 18.8 GiB
Avg memory 6.4 GiB
CPU user 830.4s
CPU sys 86.2s
Peak spill 0 B

File an issue against this benchmark runner

@rluvaton

rluvaton commented Oct 20, 2026

Copy link
Copy Markdown
Member Author

run benchmark sort
env:
BENCH_FILTER: sort_axis/sort.*.1M.High.cardinality.Unsorted..100cols

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6293178287-3645-nbcv8 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort
env:
  BENCH_FILTER: "sort_axis/sort.*.1M.High.cardinality.Unsorted..100cols"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Oct 20, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing sort-background (2b4deca) to d89a85d (merge-base) diff

Run configuration 
run benchmark sort
env:
  BENCH_FILTER: "sort_axis/sort.*.1M.High.cardinality.Unsorted..100cols"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

group                                                               HEAD                                   sort-background
-----                                                               ----                                   ---------------
sort_axis/sort i64 1M High cardinality Unsorted +100cols            1.00       2.1±0.05s        ? ?/sec    1.17       2.5±0.04s        ? ?/sec
sort_axis/sort mixed tuple 1M High cardinality Unsorted +100cols    1.00       2.2±0.08s        ? ?/sec    1.22       2.7±0.08s        ? ?/sec
sort_axis/sort utf8 view 1M High cardinality Unsorted +100cols      1.00       2.1±0.02s        ? ?/sec    1.17       2.5±0.11s        ? ?/sec

Resource Usage

sort — base (merge-base)

Metric Value
Wall time 2255.5s
Peak memory 2.3 GiB
Avg memory 1.4 GiB
CPU user 2180.8s
CPU sys 111.2s
Peak spill 0 B

sort — branch

Metric Value
Wall time 2545.6s
Peak memory 9.0 GiB
Avg memory 3.4 GiB
CPU user 2162.4s
CPU sys 84.1s
Peak spill 0 B

File an issue against this benchmark runner

@github-actions

github-actions Bot commented Oct 20, 2026 •
edited
Loading

Copy link
Copy Markdown

Thank you for opening this pull request!

Reviewer note: cargo-semver-checks reported the current version number is not SemVer-compatible with the changes in this pull request (compared against the base branch).

Details 
     Cloning apache/main
    Building datafusion-common v55.1.0 (current)
       Built [  21.462s] (current)
     Parsing datafusion-common v55.1.0 (current)
      Parsed [   0.047s] (current)
    Building datafusion-common v55.1.0 (baseline)
       Built [  21.854s] (baseline)
     Parsing datafusion-common v55.1.0 (baseline)
      Parsed [   0.047s] (baseline)
    Checking datafusion-common v55.1.0 -> v55.1.0 (no change; assume patch)
     Checked [   0.551s] 229 checks: 228 pass, 1 fail, 0 warn, 31 skip

--- failure constructible_struct_adds_field: struct exhaustively constructible through public API adds field ---

Description:
A pub struct that could be exhaustively constructed with a literal using only public API has a new pub field, breaking existing exhaustive literals.
        ref: https://doc.rust-lang.org/reference/expressions/struct-expr.html
       impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.51.0/src/lints/constructible_struct_adds_field.ron

Failed in:
  field ExecutionOptions.sort_background_merge_batches in /home/runner/work/datafusion/datafusion/datafusion/common/src/config.rs:897

     Summary semver requires new major version: 1 major and 0 minor checks failed
    Finished [  45.682s] datafusion-common
    Building datafusion-execution v55.1.0 (current)
       Built [  20.062s] (current)
     Parsing datafusion-execution v55.1.0 (current)
      Parsed [   0.021s] (current)
    Building datafusion-execution v55.1.0 (baseline)
       Built [  19.969s] (baseline)
     Parsing datafusion-execution v55.1.0 (baseline)
      Parsed [   0.021s] (baseline)
    Checking datafusion-execution v55.1.0 -> v55.1.0 (no change; assume patch)
     Checked [   0.182s] 229 checks: 229 pass, 31 skip
     Summary no semver update required
    Finished [  41.302s] datafusion-execution
    Building datafusion-physical-plan v55.1.0 (current)
       Built [  23.301s] (current)
     Parsing datafusion-physical-plan v55.1.0 (current)
      Parsed [   0.138s] (current)
    Building datafusion-physical-plan v55.1.0 (baseline)
       Built [  23.779s] (baseline)
     Parsing datafusion-physical-plan v55.1.0 (baseline)
      Parsed [   0.135s] (baseline)
    Checking datafusion-physical-plan v55.1.0 -> v55.1.0 (no change; assume patch)
     Checked [   0.486s] 229 checks: 229 pass, 31 skip
     Summary no semver update required
    Finished [  49.123s] datafusion-physical-plan
    Building datafusion-sqllogictest v55.1.0 (current)
       Built [  62.558s] (current)
     Parsing datafusion-sqllogictest v55.1.0 (current)
      Parsed [   0.012s] (current)
    Building datafusion-sqllogictest v55.1.0 (baseline)
       Built [  64.269s] (baseline)
     Parsing datafusion-sqllogictest v55.1.0 (baseline)
      Parsed [   0.013s] (baseline)
    Checking datafusion-sqllogictest v55.1.0 -> v55.1.0 (no change; assume patch)
     Checked [   0.074s] 229 checks: 229 pass, 31 skip
     Summary no semver update required
    Finished [ 129.306s] datafusion-sqllogictest

@github-actions github-actions Bot added the auto detected api change Auto detected API change label Oct 20, 2026
@codecov-commenter

codecov-commenter commented Oct 20, 2026

Copy link
Copy Markdown

Codecov Report

❌ Patch coverage is 89.76945% with 71 lines in your changes missing coverage. Please review.
✅ Project coverage is 82.94%. Comparing base (d89a85d) to head (2b4deca).
⚠️ Report is 10 commits behind head on main.

Files with missing lines Patch % Lines
datafusion/physical-plan/src/sorts/sort.rs 89.65% 13 Missing and 58 partials ⚠️
Additional details and impacted files 
@@            Coverage Diff             @@
##             main   #26527      +/-   ##
==========================================
+ Coverage   82.93%   82.94%   +0.01%     
==========================================
  Files        1154     1154              
  Lines      457933   458684     +751     
  Branches   457933   458684     +751     
==========================================
+ Hits       379786   380458     +672     
- Misses      55128    55140      +12     
- Partials    23019    23086      +67

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

@rluvaton

rluvaton commented Oct 22, 2026

Copy link
Copy Markdown
Member Author

not showed enough benefit

@rluvaton rluvaton closed this Oct 22, 2026
@rluvaton
rluvaton deleted the sort-background branch October 22, 2026 09:06
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

auto detected api change Auto detected API change common Related to common crate documentation Improvements or additions to documentation execution Related to the execution crate physical-plan Changes to the physical-plan crate sqllogictest SQL Logic Tests (.slt)

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@rluvaton @adriangbot @codecov-commenter