Dynamic work scheduling in FileStream

[alamb]

Which issue does this PR close?

• Closes Introduce morsel-driven Parquet scan #20529
• Closes Sketch out a Morselize API #20820

Rationale for this change

This PR finally enables dynamic work scheduling in the FileStream (so that if a task is done it can look at any remaining work)

This improves performance on queries that scan multiple files and the work is not balanced evenly across partitions in the plan (e.g. we have dynamic filtering that reduces work significantly)

It is the last of a sequence of several PRs:

• Rewrite FileStream in terms of Morsel API #21342
• Introduce Morselizer API, rewrite ParquetOpener to ParquetMorselizer #21327
• Split FileStreamMetrics into its own module #21340

What changes are included in this PR?

1. Add shared state across sibling FileStream's and the wiring to connect them
2. Sibling streams put their file work into a shared queue when it can be reordered
3. Add a bunch of tests sjpw

Note there are a bunch of other things that are NOT included in this PR, including

1. Trying to limit concurrent IO (this PR has the same properties as main -- up to one outstanding IO per partition)
2. Trying to issue multiple IOs by the same partition (aka to interleave IO and CPU work more)
3. Splitting files into smaller units (e.g. across row groups)

As @Dandandan proposes below, I expect we can work on those changes as follow on PRs.

Are these changes tested?

Yes by existing functional and benchmark tests, as well as new functional tests

Are there any user-facing changes?

Yes, faster performance (see benchmarks): #21351 (comment)

[Dandandan]

🤖 Benchmark completed (GKE) | trigger

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

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 alamb_reschedule_io
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                   alamb_reschedule_io ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.30 / 4.56 ±6.37 / 17.29 ms │          1.31 / 4.55 ±6.35 / 17.25 ms │     no change │
│ QQuery 1  │        14.65 / 15.05 ±0.29 / 15.56 ms │        12.00 / 12.42 ±0.27 / 12.83 ms │ +1.21x faster │
│ QQuery 2  │        44.23 / 44.61 ±0.42 / 45.39 ms │        35.97 / 36.54 ±0.31 / 36.90 ms │ +1.22x faster │
│ QQuery 3  │        43.10 / 44.46 ±1.35 / 46.69 ms │        34.41 / 35.17 ±0.70 / 36.12 ms │ +1.26x faster │
│ QQuery 4  │     293.78 / 305.01 ±7.65 / 316.85 ms │     260.16 / 263.55 ±4.78 / 272.96 ms │ +1.16x faster │
│ QQuery 5  │     348.16 / 352.47 ±4.23 / 358.82 ms │     302.86 / 308.72 ±5.56 / 319.27 ms │ +1.14x faster │
│ QQuery 6  │           5.80 / 6.57 ±0.63 / 7.63 ms │           6.62 / 7.86 ±0.78 / 8.75 ms │  1.20x slower │
│ QQuery 7  │        17.11 / 17.74 ±0.66 / 18.58 ms │        13.90 / 14.35 ±0.39 / 14.93 ms │ +1.24x faster │
│ QQuery 8  │     420.14 / 429.93 ±8.45 / 444.77 ms │    378.42 / 390.37 ±11.88 / 408.19 ms │ +1.10x faster │
│ QQuery 9  │     663.03 / 672.84 ±5.72 / 680.54 ms │    521.39 / 531.97 ±10.80 / 552.34 ms │ +1.26x faster │
│ QQuery 10 │        92.14 / 94.93 ±2.52 / 99.25 ms │        76.47 / 77.79 ±1.46 / 80.45 ms │ +1.22x faster │
│ QQuery 11 │     104.15 / 105.38 ±0.69 / 106.30 ms │        86.41 / 87.26 ±0.97 / 89.08 ms │ +1.21x faster │
│ QQuery 12 │     357.65 / 366.42 ±6.72 / 373.88 ms │     297.51 / 304.37 ±4.38 / 309.09 ms │ +1.20x faster │
│ QQuery 13 │     458.02 / 470.34 ±9.64 / 480.11 ms │    428.91 / 443.12 ±10.92 / 458.50 ms │ +1.06x faster │
│ QQuery 14 │     354.42 / 359.89 ±4.07 / 367.09 ms │     305.02 / 314.80 ±6.65 / 325.44 ms │ +1.14x faster │
│ QQuery 15 │    370.27 / 382.31 ±15.57 / 412.33 ms │    332.09 / 345.10 ±15.11 / 373.91 ms │ +1.11x faster │
│ QQuery 16 │    734.79 / 751.32 ±12.64 / 771.68 ms │    690.14 / 723.75 ±56.61 / 836.78 ms │     no change │
│ QQuery 17 │    721.23 / 731.36 ±10.62 / 751.20 ms │    686.92 / 718.61 ±57.67 / 833.81 ms │     no change │
│ QQuery 18 │ 1418.87 / 1491.07 ±39.56 / 1534.75 ms │ 1378.18 / 1409.22 ±49.85 / 1508.29 ms │ +1.06x faster │
│ QQuery 19 │        36.52 / 38.08 ±1.80 / 40.51 ms │      32.21 / 62.83 ±37.79 / 137.37 ms │  1.65x slower │
│ QQuery 20 │    723.14 / 740.27 ±16.11 / 760.99 ms │    513.72 / 527.72 ±10.71 / 544.27 ms │ +1.40x faster │
│ QQuery 21 │     767.80 / 774.94 ±6.76 / 787.66 ms │     597.68 / 606.11 ±7.93 / 618.51 ms │ +1.28x faster │
│ QQuery 22 │  1131.51 / 1137.57 ±7.04 / 1151.24 ms │ 1062.84 / 1072.76 ±10.36 / 1091.17 ms │ +1.06x faster │
│ QQuery 23 │ 3068.54 / 3099.60 ±19.20 / 3116.41 ms │ 3302.70 / 3323.50 ±16.22 / 3352.12 ms │  1.07x slower │
│ QQuery 24 │     101.01 / 104.87 ±2.81 / 109.76 ms │        43.34 / 46.80 ±3.25 / 52.18 ms │ +2.24x faster │
│ QQuery 25 │     140.64 / 141.29 ±0.67 / 142.40 ms │     114.94 / 116.41 ±1.24 / 118.08 ms │ +1.21x faster │
│ QQuery 26 │     102.97 / 104.39 ±0.80 / 105.19 ms │        43.89 / 45.52 ±1.57 / 47.70 ms │ +2.29x faster │
│ QQuery 27 │     852.42 / 859.56 ±7.14 / 871.93 ms │     673.53 / 683.34 ±5.19 / 688.03 ms │ +1.26x faster │
│ QQuery 28 │ 7750.35 / 7784.12 ±20.30 / 7804.81 ms │ 6987.03 / 7070.54 ±44.85 / 7114.20 ms │ +1.10x faster │
│ QQuery 29 │        52.98 / 55.46 ±4.12 / 63.62 ms │       43.51 / 55.84 ±19.75 / 95.00 ms │     no change │
│ QQuery 30 │     364.88 / 374.97 ±7.27 / 384.57 ms │    332.34 / 344.98 ±13.87 / 371.55 ms │ +1.09x faster │
│ QQuery 31 │    372.94 / 390.15 ±14.22 / 416.32 ms │    335.38 / 350.66 ±13.06 / 374.32 ms │ +1.11x faster │
│ QQuery 32 │ 1046.43 / 1088.88 ±41.29 / 1149.35 ms │ 1038.82 / 1060.23 ±30.52 / 1120.27 ms │     no change │
│ QQuery 33 │  1488.55 / 1499.96 ±7.93 / 1512.60 ms │  1459.04 / 1470.59 ±7.68 / 1480.30 ms │     no change │
│ QQuery 34 │ 1488.46 / 1500.84 ±11.87 / 1519.15 ms │ 1447.16 / 1472.97 ±18.26 / 1500.69 ms │     no change │
│ QQuery 35 │     402.87 / 405.09 ±2.05 / 408.76 ms │     322.39 / 325.27 ±4.63 / 334.50 ms │ +1.25x faster │
│ QQuery 36 │     117.00 / 122.22 ±2.88 / 125.83 ms │        70.69 / 72.64 ±1.35 / 74.20 ms │ +1.68x faster │
│ QQuery 37 │        50.38 / 51.37 ±0.73 / 52.43 ms │        38.61 / 39.88 ±0.92 / 41.10 ms │ +1.29x faster │
│ QQuery 38 │        76.42 / 77.64 ±0.93 / 78.65 ms │        40.72 / 42.41 ±1.03 / 43.59 ms │ +1.83x faster │
│ QQuery 39 │     222.54 / 226.06 ±2.44 / 229.01 ms │     130.19 / 134.34 ±3.95 / 140.15 ms │ +1.68x faster │
│ QQuery 40 │        22.89 / 26.76 ±2.41 / 30.20 ms │        17.75 / 20.33 ±2.11 / 22.87 ms │ +1.32x faster │
│ QQuery 41 │        19.87 / 21.12 ±0.68 / 21.88 ms │        15.20 / 15.73 ±0.36 / 16.07 ms │ +1.34x faster │
│ QQuery 42 │        20.10 / 20.26 ±0.15 / 20.52 ms │        13.64 / 14.18 ±0.41 / 14.70 ms │ +1.43x faster │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                  │ 27291.72ms │
│ Total Time (alamb_reschedule_io)   │ 25005.09ms │
│ Average Time (HEAD)                │   634.69ms │
│ Average Time (alamb_reschedule_io) │   581.51ms │
│ Queries Faster                     │         33 │
│ Queries Slower                     │          3 │
│ Queries with No Change             │          7 │
│ Queries with Failure               │          0 │
└────────────────────────────────────┴────────────┘

Resource Usage
clickbench_partitioned — base (merge-base)

Metric Value
Wall time 137.6s
Peak memory 40.4 GiB
Avg memory 31.6 GiB
CPU user 1299.0s
CPU sys 87.2s
Peak spill 0 B
clickbench_partitioned — branch

Metric Value
Wall time 126.2s
Peak memory 41.5 GiB
Avg memory 28.2 GiB
CPU user 1313.5s
CPU sys 87.9s
Peak spill 0 B
File an issue against this benchmark runner

It's happening!

[alamb]

It's happening!

Yes! Finally -- and no regressions this time!

I have a pile of stacked PRs:

• Introduce Morselizer API, rewrite ParquetOpener to ParquetMorselizer #21327
• Rewrite FileStream in terms of Morsel API #21342

Now that I have some evidence that this PR make them faster, I'll go back and get them ready to review

[alamb]

[comphead]

I asked Claude to write some tests on Shared mode and it found some potential concerns

 When a partition hits its row limit, scan_state.rs:175 runs:                                                                                                                                                                                                                                                     
                                                                                                                                                                                                                                                                                                                 
  let done = 1 + self.work_source.len();                                                                                                                                                                                                                                                                           
  self.metrics.files_processed.add(done);                                                                                                                                                                                                                                                                          
                                                                                                                                                                                                                                                                                                                   
  This was correct before work stealing — work_source.len() returned the count of files owned exclusively by this stream. Hitting the limit meant those files would never be opened, so marking them all as "processed" (skipped) was accurate.                                                                    
                                                                                                                                                                                                                                                                                                                   
  With WorkSource::Shared, len() returns the count of files still in the shared queue. Those files aren't skipped — sibling streams will pop and process them. So they get counted twice:                                                                                                                          
                                                            
  1. The limiting partition adds them to files_processed (treating them as skipped)                                                                                                                                                                                                                                
  2. The sibling that actually processes them adds 1 each as it finishes them
                                                                                                                                                                                                                                                                                                                   
  In the test: 4 total files, partition 1 hits the limit and counts 1 + 3 = 4, then partition 0 processes the remaining 3. Result: files_processed = 7 instead of 4.                                                                                                                                               
                                                                                                                                                                                                                                                                                                                   
  The fix: when hitting a limit with a shared work source, only count the current file (done = 1), not the shared queue remainder.  
                                                                                                                                                                                                                                                                                                                   
  The fix would be in scan_state.rs:175 — for the WorkSource::Shared case, done should be 1 (only the current file), not 1 + shared_queue.len().     

[alamb]

I asked Claude to write some tests on Shared mode and it found some potential concerns

I think this is a real bug (for metric reporting). I have pushed a fix for it. Thank you @comphead

[Dandandan]

I think it looks correct and the changes also look quite simple to me (after the refactoring PRs).

[Dandandan]

run benchmarks

[Dandandan]

run benchmarks

env:
  DATAFUSION_EXECUTION_PARQUET_PUSHDOWN_FILTERS: true
  DATAFUSION_EXECUTION_PARQUET_REORDER_FILTERS: true

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4266775584-1413-xbz4p 6.12.55+ #1 SMP Sun Feb 1 08:59:41 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 alamb/reschedule_io (323a5ca) to 7731130 (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4266775584-1414-6prwj 6.12.55+ #1 SMP Sun Feb 1 08:59:41 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 alamb/reschedule_io (323a5ca) to 7731130 (merge-base) diff using: tpcds
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4266775584-1415-5n9jg 6.12.55+ #1 SMP Sun Feb 1 08:59:41 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 alamb/reschedule_io (323a5ca) to 7731130 (merge-base) diff using: tpch
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4266791680-1418-5tndt 6.12.55+ #1 SMP Sun Feb 1 08:59:41 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 alamb/reschedule_io (323a5ca) to 7731130 (merge-base) diff using: tpch
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4266791680-1417-khxln 6.12.55+ #1 SMP Sun Feb 1 08:59:41 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 alamb/reschedule_io (323a5ca) to 7731130 (merge-base) diff using: tpcds
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4266791680-1416-2sk4z 6.12.55+ #1 SMP Sun Feb 1 08:59:41 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 alamb/reschedule_io (323a5ca) to 7731130 (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

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

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 alamb_reschedule_io
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                   alamb_reschedule_io ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.20 / 4.41 ±6.36 / 17.13 ms │          1.19 / 4.41 ±6.36 / 17.14 ms │     no change │
│ QQuery 1  │        15.45 / 15.62 ±0.13 / 15.83 ms │        12.04 / 12.50 ±0.28 / 12.90 ms │ +1.25x faster │
│ QQuery 2  │        43.35 / 43.78 ±0.33 / 44.19 ms │        36.37 / 36.94 ±0.51 / 37.86 ms │ +1.19x faster │
│ QQuery 3  │        39.02 / 39.74 ±1.02 / 41.74 ms │        33.89 / 34.96 ±1.07 / 36.39 ms │ +1.14x faster │
│ QQuery 4  │    277.09 / 288.53 ±12.16 / 311.57 ms │    250.06 / 260.25 ±10.85 / 281.11 ms │ +1.11x faster │
│ QQuery 5  │     335.35 / 340.47 ±4.75 / 347.31 ms │     297.67 / 302.70 ±4.49 / 308.39 ms │ +1.12x faster │
│ QQuery 6  │          6.27 / 8.49 ±2.45 / 12.00 ms │          5.72 / 7.39 ±2.00 / 11.25 ms │ +1.15x faster │
│ QQuery 7  │        21.49 / 22.06 ±0.49 / 22.95 ms │        15.79 / 16.08 ±0.28 / 16.54 ms │ +1.37x faster │
│ QQuery 8  │     403.43 / 408.73 ±4.15 / 415.98 ms │     359.37 / 367.80 ±7.93 / 382.02 ms │ +1.11x faster │
│ QQuery 9  │     622.67 / 631.51 ±6.61 / 641.33 ms │     495.34 / 509.66 ±8.97 / 522.89 ms │ +1.24x faster │
│ QQuery 10 │     116.45 / 119.36 ±2.34 / 123.24 ms │      98.73 / 100.88 ±2.41 / 105.55 ms │ +1.18x faster │
│ QQuery 11 │     129.04 / 130.35 ±0.99 / 131.75 ms │     108.59 / 109.67 ±0.68 / 110.41 ms │ +1.19x faster │
│ QQuery 12 │     368.27 / 375.98 ±5.87 / 382.83 ms │     329.72 / 336.06 ±7.78 / 351.14 ms │ +1.12x faster │
│ QQuery 13 │    477.83 / 502.57 ±22.41 / 533.48 ms │     463.69 / 470.26 ±6.29 / 480.77 ms │ +1.07x faster │
│ QQuery 14 │     373.47 / 380.17 ±4.93 / 388.57 ms │     337.86 / 343.32 ±4.57 / 350.81 ms │ +1.11x faster │
│ QQuery 15 │    346.39 / 367.67 ±18.34 / 396.05 ms │    324.21 / 338.37 ±16.87 / 369.67 ms │ +1.09x faster │
│ QQuery 16 │     711.66 / 716.94 ±7.10 / 730.59 ms │     684.81 / 688.78 ±3.14 / 694.06 ms │     no change │
│ QQuery 17 │     705.03 / 712.55 ±5.48 / 720.32 ms │     670.37 / 680.08 ±5.14 / 685.05 ms │     no change │
│ QQuery 18 │ 1382.27 / 1472.54 ±48.23 / 1523.29 ms │ 1342.68 / 1469.83 ±68.97 / 1538.78 ms │     no change │
│ QQuery 19 │        36.20 / 41.18 ±7.79 / 56.69 ms │        30.45 / 32.11 ±2.68 / 37.39 ms │ +1.28x faster │
│ QQuery 20 │    711.01 / 734.08 ±20.59 / 761.19 ms │    513.35 / 523.27 ±10.01 / 541.18 ms │ +1.40x faster │
│ QQuery 21 │    713.78 / 733.18 ±11.23 / 743.85 ms │     576.65 / 580.50 ±3.25 / 586.16 ms │ +1.26x faster │
│ QQuery 22 │ 1182.69 / 1198.17 ±24.04 / 1245.97 ms │    942.75 / 959.22 ±11.64 / 977.58 ms │ +1.25x faster │
│ QQuery 23 │     258.67 / 269.82 ±7.09 / 280.41 ms │     122.32 / 130.62 ±6.77 / 139.39 ms │ +2.07x faster │
│ QQuery 24 │        91.90 / 94.66 ±1.99 / 97.90 ms │        44.51 / 45.96 ±1.19 / 47.39 ms │ +2.06x faster │
│ QQuery 25 │     175.99 / 179.12 ±2.37 / 183.28 ms │     143.88 / 145.31 ±1.32 / 147.77 ms │ +1.23x faster │
│ QQuery 26 │     110.12 / 112.31 ±2.16 / 115.27 ms │        59.42 / 63.16 ±2.10 / 65.49 ms │ +1.78x faster │
│ QQuery 27 │    954.54 / 972.14 ±15.14 / 996.06 ms │     712.70 / 725.20 ±9.35 / 737.80 ms │ +1.34x faster │
│ QQuery 28 │ 3363.83 / 3405.74 ±22.31 / 3428.02 ms │ 3101.44 / 3135.40 ±29.68 / 3182.63 ms │ +1.09x faster │
│ QQuery 29 │        49.90 / 57.00 ±8.90 / 74.31 ms │        43.50 / 47.80 ±4.66 / 55.84 ms │ +1.19x faster │
│ QQuery 30 │    358.79 / 370.85 ±10.70 / 390.56 ms │     331.65 / 338.27 ±4.84 / 346.46 ms │ +1.10x faster │
│ QQuery 31 │     365.03 / 376.43 ±9.80 / 388.54 ms │     326.37 / 336.33 ±6.32 / 344.47 ms │ +1.12x faster │
│ QQuery 32 │ 1151.16 / 1207.90 ±66.95 / 1329.94 ms │ 1200.05 / 1269.48 ±38.71 / 1312.73 ms │  1.05x slower │
│ QQuery 33 │ 1444.73 / 1482.38 ±37.24 / 1552.68 ms │ 1428.91 / 1443.61 ±18.30 / 1477.76 ms │     no change │
│ QQuery 34 │ 1441.71 / 1474.94 ±21.10 / 1496.78 ms │  1444.56 / 1453.39 ±7.93 / 1465.49 ms │     no change │
│ QQuery 35 │     376.92 / 383.56 ±3.90 / 389.17 ms │     309.17 / 311.39 ±2.56 / 316.25 ms │ +1.23x faster │
│ QQuery 36 │      93.50 / 103.97 ±5.54 / 108.71 ms │        66.30 / 68.56 ±2.02 / 72.00 ms │ +1.52x faster │
│ QQuery 37 │        52.48 / 55.58 ±2.51 / 58.91 ms │        43.61 / 44.67 ±0.67 / 45.40 ms │ +1.24x faster │
│ QQuery 38 │        52.62 / 57.62 ±2.83 / 60.80 ms │        37.22 / 40.31 ±2.06 / 43.42 ms │ +1.43x faster │
│ QQuery 39 │     171.26 / 177.05 ±4.76 / 184.17 ms │     113.14 / 121.23 ±4.26 / 125.85 ms │ +1.46x faster │
│ QQuery 40 │        26.78 / 30.54 ±2.67 / 33.81 ms │        20.86 / 22.47 ±0.92 / 23.61 ms │ +1.36x faster │
│ QQuery 41 │        25.53 / 27.73 ±1.37 / 29.32 ms │        18.90 / 20.23 ±1.50 / 23.04 ms │ +1.37x faster │
│ QQuery 42 │        21.82 / 22.07 ±0.17 / 22.30 ms │        14.53 / 15.08 ±0.36 / 15.61 ms │ +1.46x faster │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                  │ 20149.48ms │
│ Total Time (alamb_reschedule_io)   │ 17963.47ms │
│ Average Time (HEAD)                │   468.59ms │
│ Average Time (alamb_reschedule_io) │   417.76ms │
│ Queries Faster                     │         36 │
│ Queries Slower                     │          1 │
│ Queries with No Change             │          6 │
│ Queries with Failure               │          0 │
└────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	101.7s
Peak memory	37.4 GiB
Avg memory	27.2 GiB
CPU user	914.3s
CPU sys	98.4s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	91.0s
Peak memory	42.0 GiB
Avg memory	29.3 GiB
CPU user	914.8s
CPU sys	94.7s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

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

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 alamb_reschedule_io
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                   alamb_reschedule_io ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.21 / 4.51 ±6.42 / 17.35 ms │          1.18 / 4.46 ±6.40 / 17.26 ms │     no change │
│ QQuery 1  │        15.06 / 15.51 ±0.29 / 15.94 ms │        11.83 / 12.32 ±0.25 / 12.48 ms │ +1.26x faster │
│ QQuery 2  │        44.77 / 45.32 ±0.36 / 45.87 ms │        36.16 / 36.37 ±0.18 / 36.66 ms │ +1.25x faster │
│ QQuery 3  │        40.43 / 41.64 ±1.42 / 43.88 ms │        30.79 / 31.38 ±0.36 / 31.78 ms │ +1.33x faster │
│ QQuery 4  │     290.10 / 296.61 ±6.45 / 304.54 ms │     252.54 / 258.82 ±5.50 / 265.87 ms │ +1.15x faster │
│ QQuery 5  │     347.94 / 351.60 ±2.96 / 355.36 ms │     302.53 / 306.74 ±2.17 / 308.49 ms │ +1.15x faster │
│ QQuery 6  │           4.98 / 5.85 ±0.60 / 6.81 ms │           5.79 / 6.70 ±0.88 / 8.37 ms │  1.14x slower │
│ QQuery 7  │        17.84 / 18.09 ±0.31 / 18.68 ms │        13.72 / 15.72 ±3.50 / 22.69 ms │ +1.15x faster │
│ QQuery 8  │     433.11 / 436.74 ±2.91 / 440.16 ms │     363.98 / 370.12 ±7.91 / 385.47 ms │ +1.18x faster │
│ QQuery 9  │    671.55 / 685.15 ±13.69 / 705.13 ms │     510.02 / 518.50 ±6.03 / 526.08 ms │ +1.32x faster │
│ QQuery 10 │        92.51 / 95.98 ±2.67 / 99.79 ms │        77.76 / 79.57 ±2.57 / 84.59 ms │ +1.21x faster │
│ QQuery 11 │     105.46 / 107.53 ±1.51 / 108.92 ms │        87.61 / 89.39 ±1.05 / 90.44 ms │ +1.20x faster │
│ QQuery 12 │     342.24 / 344.29 ±1.97 / 347.26 ms │     294.05 / 303.01 ±8.70 / 317.59 ms │ +1.14x faster │
│ QQuery 13 │    461.32 / 475.59 ±12.08 / 492.95 ms │    430.77 / 452.49 ±22.43 / 495.45 ms │     no change │
│ QQuery 14 │     349.20 / 351.37 ±2.04 / 355.26 ms │     308.17 / 311.54 ±2.33 / 315.19 ms │ +1.13x faster │
│ QQuery 15 │    347.27 / 364.14 ±15.38 / 389.85 ms │    312.82 / 331.92 ±18.42 / 365.43 ms │ +1.10x faster │
│ QQuery 16 │     706.04 / 719.63 ±9.48 / 734.25 ms │     681.81 / 691.46 ±8.03 / 705.69 ms │     no change │
│ QQuery 17 │     704.76 / 710.08 ±4.04 / 714.77 ms │     690.34 / 695.20 ±3.53 / 700.35 ms │     no change │
│ QQuery 18 │ 1400.64 / 1482.31 ±43.16 / 1521.24 ms │ 1348.05 / 1369.28 ±12.84 / 1382.45 ms │ +1.08x faster │
│ QQuery 19 │        37.82 / 38.54 ±0.79 / 39.94 ms │        29.25 / 31.41 ±2.04 / 35.15 ms │ +1.23x faster │
│ QQuery 20 │    718.69 / 736.09 ±14.95 / 756.97 ms │     510.85 / 515.75 ±5.38 / 525.77 ms │ +1.43x faster │
│ QQuery 21 │     760.02 / 764.01 ±6.11 / 776.11 ms │     590.87 / 593.20 ±1.61 / 595.69 ms │ +1.29x faster │
│ QQuery 22 │  1127.59 / 1133.75 ±7.46 / 1148.21 ms │  1048.15 / 1050.37 ±2.29 / 1054.49 ms │ +1.08x faster │
│ QQuery 23 │ 3071.26 / 3099.27 ±23.81 / 3142.02 ms │  3315.87 / 3326.69 ±7.25 / 3338.15 ms │  1.07x slower │
│ QQuery 24 │      98.19 / 102.45 ±5.01 / 112.01 ms │        44.75 / 46.27 ±1.36 / 48.32 ms │ +2.21x faster │
│ QQuery 25 │     140.35 / 142.98 ±2.33 / 146.63 ms │     114.07 / 116.09 ±1.29 / 117.52 ms │ +1.23x faster │
│ QQuery 26 │     103.52 / 104.33 ±0.66 / 105.25 ms │        43.10 / 44.43 ±1.69 / 47.29 ms │ +2.35x faster │
│ QQuery 27 │     846.11 / 851.64 ±4.15 / 858.73 ms │     658.99 / 666.08 ±3.81 / 669.13 ms │ +1.28x faster │
│ QQuery 28 │ 3240.63 / 3266.62 ±16.79 / 3283.65 ms │ 2996.24 / 3028.52 ±19.67 / 3046.78 ms │ +1.08x faster │
│ QQuery 29 │        51.14 / 53.89 ±2.46 / 58.21 ms │        44.30 / 48.63 ±3.99 / 54.79 ms │ +1.11x faster │
│ QQuery 30 │     358.16 / 365.59 ±7.11 / 377.55 ms │     336.01 / 341.62 ±5.88 / 352.60 ms │ +1.07x faster │
│ QQuery 31 │     375.91 / 384.62 ±8.26 / 394.86 ms │     332.04 / 342.14 ±6.26 / 349.75 ms │ +1.12x faster │
│ QQuery 32 │ 1039.53 / 1081.16 ±53.10 / 1179.57 ms │  1025.47 / 1040.28 ±9.44 / 1055.30 ms │     no change │
│ QQuery 33 │ 1455.36 / 1470.57 ±10.22 / 1482.86 ms │  1467.32 / 1476.46 ±7.15 / 1488.74 ms │     no change │
│ QQuery 34 │ 1469.24 / 1488.26 ±12.82 / 1509.11 ms │ 1478.51 / 1518.74 ±22.34 / 1543.99 ms │     no change │
│ QQuery 35 │     388.12 / 394.26 ±3.49 / 398.75 ms │     321.75 / 326.54 ±4.52 / 334.40 ms │ +1.21x faster │
│ QQuery 36 │     117.64 / 120.01 ±2.45 / 124.07 ms │        64.65 / 67.92 ±1.98 / 70.05 ms │ +1.77x faster │
│ QQuery 37 │        48.39 / 50.27 ±1.45 / 52.88 ms │        37.85 / 39.46 ±1.84 / 42.93 ms │ +1.27x faster │
│ QQuery 38 │        75.51 / 77.67 ±1.13 / 78.83 ms │        43.12 / 43.58 ±0.26 / 43.90 ms │ +1.78x faster │
│ QQuery 39 │     205.95 / 217.49 ±8.52 / 230.07 ms │     126.07 / 134.11 ±5.72 / 142.29 ms │ +1.62x faster │
│ QQuery 40 │        24.82 / 26.61 ±1.69 / 29.71 ms │        19.36 / 21.20 ±2.08 / 24.57 ms │ +1.26x faster │
│ QQuery 41 │        20.66 / 21.90 ±1.08 / 23.77 ms │        14.52 / 15.63 ±0.63 / 16.41 ms │ +1.40x faster │
│ QQuery 42 │        20.19 / 20.40 ±0.27 / 20.91 ms │        13.72 / 14.27 ±0.50 / 15.10 ms │ +1.43x faster │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                  │ 22564.32ms │
│ Total Time (alamb_reschedule_io)   │ 20734.35ms │
│ Average Time (HEAD)                │   524.75ms │
│ Average Time (alamb_reschedule_io) │   482.19ms │
│ Queries Faster                     │         34 │
│ Queries Slower                     │          2 │
│ Queries with No Change             │          7 │
│ Queries with Failure               │          0 │
└────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	113.8s
Peak memory	37.5 GiB
Avg memory	26.9 GiB
CPU user	1074.2s
CPU sys	85.2s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	104.6s
Peak memory	38.1 GiB
Avg memory	32.2 GiB
CPU user	1090.1s
CPU sys	78.6s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

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

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 alamb_reschedule_io
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃                      alamb_reschedule_io ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │              7.06 / 7.45 ±0.66 / 8.77 ms │              7.05 / 7.47 ±0.71 / 8.89 ms │     no change │
│ QQuery 2  │        146.21 / 146.56 ±0.25 / 146.99 ms │        148.37 / 149.21 ±0.53 / 149.91 ms │     no change │
│ QQuery 3  │        115.47 / 116.08 ±0.91 / 117.87 ms │        114.28 / 115.04 ±0.63 / 115.91 ms │     no change │
│ QQuery 4  │    1434.70 / 1464.26 ±18.15 / 1486.16 ms │    1425.39 / 1456.36 ±21.83 / 1486.26 ms │     no change │
│ QQuery 5  │        175.39 / 179.27 ±2.26 / 182.06 ms │        173.25 / 175.11 ±1.45 / 177.64 ms │     no change │
│ QQuery 6  │       907.35 / 930.61 ±24.14 / 976.35 ms │        885.02 / 891.70 ±4.79 / 896.55 ms │     no change │
│ QQuery 7  │        351.05 / 353.63 ±2.25 / 356.69 ms │        346.98 / 349.88 ±2.00 / 352.72 ms │     no change │
│ QQuery 8  │        120.85 / 122.54 ±1.75 / 125.28 ms │        116.49 / 118.24 ±1.31 / 120.41 ms │     no change │
│ QQuery 9  │        102.71 / 105.53 ±2.84 / 109.04 ms │        102.03 / 106.84 ±2.56 / 109.21 ms │     no change │
│ QQuery 10 │        109.91 / 111.62 ±1.53 / 114.42 ms │        109.04 / 110.39 ±1.07 / 112.10 ms │     no change │
│ QQuery 11 │    1041.44 / 1051.90 ±10.87 / 1066.52 ms │     1030.40 / 1036.96 ±5.82 / 1045.53 ms │     no change │
│ QQuery 12 │           45.84 / 47.88 ±1.83 / 50.78 ms │           47.98 / 49.13 ±1.59 / 52.21 ms │     no change │
│ QQuery 13 │        413.39 / 416.95 ±3.34 / 422.11 ms │        411.03 / 412.94 ±2.09 / 416.95 ms │     no change │
│ QQuery 14 │     1012.26 / 1026.35 ±8.48 / 1038.09 ms │     1003.58 / 1009.99 ±5.66 / 1018.32 ms │     no change │
│ QQuery 15 │           16.61 / 18.28 ±1.62 / 20.82 ms │           15.79 / 16.81 ±0.78 / 17.81 ms │ +1.09x faster │
│ QQuery 16 │              7.51 / 8.09 ±0.81 / 9.68 ms │             7.42 / 8.18 ±1.17 / 10.48 ms │     no change │
│ QQuery 17 │        232.47 / 236.33 ±2.16 / 238.45 ms │        235.49 / 236.22 ±0.80 / 237.71 ms │     no change │
│ QQuery 18 │        130.55 / 131.30 ±0.70 / 132.59 ms │        128.74 / 130.56 ±1.25 / 131.78 ms │     no change │
│ QQuery 19 │        159.07 / 162.26 ±2.20 / 165.36 ms │        157.83 / 159.14 ±0.84 / 160.31 ms │     no change │
│ QQuery 20 │           13.96 / 15.05 ±0.81 / 16.36 ms │           13.75 / 14.93 ±0.79 / 16.05 ms │     no change │
│ QQuery 21 │           20.39 / 21.09 ±0.40 / 21.59 ms │           19.56 / 20.37 ±0.54 / 21.19 ms │     no change │
│ QQuery 22 │        514.36 / 518.04 ±3.84 / 523.86 ms │        493.44 / 497.36 ±2.67 / 500.89 ms │     no change │
│ QQuery 23 │        922.92 / 935.43 ±9.25 / 946.94 ms │        908.02 / 913.70 ±5.95 / 924.56 ms │     no change │
│ QQuery 24 │        390.81 / 395.75 ±3.51 / 400.71 ms │        387.73 / 391.70 ±4.89 / 400.85 ms │     no change │
│ QQuery 25 │        347.39 / 351.62 ±2.76 / 355.15 ms │        346.19 / 348.06 ±1.69 / 351.09 ms │     no change │
│ QQuery 26 │           81.05 / 83.58 ±1.74 / 85.74 ms │           80.14 / 83.11 ±2.18 / 86.45 ms │     no change │
│ QQuery 27 │              7.09 / 7.17 ±0.11 / 7.38 ms │              6.96 / 7.18 ±0.17 / 7.48 ms │     no change │
│ QQuery 28 │       150.17 / 158.60 ±14.24 / 187.01 ms │        150.09 / 150.64 ±0.57 / 151.60 ms │ +1.05x faster │
│ QQuery 29 │        289.45 / 290.87 ±1.59 / 293.84 ms │        285.05 / 286.46 ±1.11 / 288.41 ms │     no change │
│ QQuery 30 │           43.27 / 45.73 ±1.76 / 48.59 ms │           43.97 / 45.07 ±0.74 / 46.29 ms │     no change │
│ QQuery 31 │        171.49 / 176.30 ±3.01 / 180.69 ms │        170.78 / 172.67 ±2.09 / 176.60 ms │     no change │
│ QQuery 32 │           58.17 / 59.28 ±1.20 / 61.19 ms │         57.35 / 67.25 ±18.73 / 104.70 ms │  1.13x slower │
│ QQuery 33 │        142.71 / 144.76 ±1.62 / 147.07 ms │        140.87 / 142.01 ±0.96 / 143.38 ms │     no change │
│ QQuery 34 │              7.37 / 7.87 ±0.51 / 8.83 ms │              6.96 / 7.27 ±0.18 / 7.50 ms │ +1.08x faster │
│ QQuery 35 │        110.18 / 112.08 ±1.45 / 113.92 ms │        106.79 / 110.23 ±2.05 / 113.25 ms │     no change │
│ QQuery 36 │              6.83 / 7.36 ±0.86 / 9.07 ms │              6.82 / 7.03 ±0.14 / 7.18 ms │     no change │
│ QQuery 37 │              8.67 / 9.00 ±0.25 / 9.40 ms │             8.52 / 9.24 ±0.89 / 10.97 ms │     no change │
│ QQuery 38 │           87.69 / 91.90 ±4.22 / 99.54 ms │           88.64 / 90.18 ±1.70 / 92.98 ms │     no change │
│ QQuery 39 │        131.16 / 134.60 ±1.82 / 136.51 ms │        125.93 / 129.17 ±2.13 / 131.96 ms │     no change │
│ QQuery 40 │        109.09 / 118.51 ±8.88 / 135.25 ms │        110.79 / 117.93 ±6.54 / 129.47 ms │     no change │
│ QQuery 41 │           15.56 / 16.72 ±0.98 / 18.53 ms │           14.45 / 16.57 ±2.12 / 20.58 ms │     no change │
│ QQuery 42 │        111.09 / 112.43 ±1.08 / 114.14 ms │        109.32 / 110.56 ±0.66 / 111.20 ms │     no change │
│ QQuery 43 │              6.34 / 6.42 ±0.09 / 6.58 ms │              6.26 / 6.48 ±0.28 / 7.03 ms │     no change │
│ QQuery 44 │           12.30 / 12.46 ±0.14 / 12.66 ms │           12.12 / 12.64 ±0.60 / 13.79 ms │     no change │
│ QQuery 45 │           51.68 / 53.37 ±0.91 / 54.22 ms │           50.97 / 53.31 ±1.51 / 55.10 ms │     no change │
│ QQuery 46 │              8.54 / 8.97 ±0.35 / 9.44 ms │              8.58 / 8.92 ±0.19 / 9.14 ms │     no change │
│ QQuery 47 │        798.85 / 804.00 ±6.26 / 816.25 ms │        750.46 / 761.09 ±7.19 / 770.67 ms │ +1.06x faster │
│ QQuery 48 │        293.82 / 295.35 ±1.09 / 296.93 ms │        291.20 / 294.98 ±3.23 / 299.85 ms │     no change │
│ QQuery 49 │        252.33 / 256.90 ±3.87 / 261.65 ms │        254.51 / 256.32 ±1.19 / 257.95 ms │     no change │
│ QQuery 50 │        229.77 / 237.34 ±4.81 / 244.61 ms │        224.14 / 229.91 ±7.59 / 243.98 ms │     no change │
│ QQuery 51 │        185.43 / 186.91 ±0.92 / 188.02 ms │        181.30 / 183.19 ±1.69 / 185.71 ms │     no change │
│ QQuery 52 │        109.40 / 112.53 ±2.84 / 117.39 ms │        108.41 / 110.67 ±1.43 / 112.54 ms │     no change │
│ QQuery 53 │        104.41 / 107.16 ±3.76 / 114.39 ms │        103.72 / 105.33 ±1.12 / 106.77 ms │     no change │
│ QQuery 54 │        149.45 / 150.86 ±0.94 / 152.05 ms │        147.16 / 148.99 ±1.81 / 151.68 ms │     no change │
│ QQuery 55 │        108.56 / 111.98 ±2.51 / 116.07 ms │        107.62 / 109.26 ±1.55 / 112.09 ms │     no change │
│ QQuery 56 │        144.59 / 145.87 ±1.32 / 147.96 ms │        141.85 / 143.45 ±1.62 / 146.56 ms │     no change │
│ QQuery 57 │        179.73 / 180.59 ±0.64 / 181.24 ms │        175.27 / 177.54 ±1.49 / 179.36 ms │     no change │
│ QQuery 58 │       304.81 / 315.48 ±14.27 / 342.84 ms │        287.77 / 298.56 ±9.31 / 310.10 ms │ +1.06x faster │
│ QQuery 59 │        202.93 / 204.57 ±1.53 / 206.90 ms │        198.95 / 201.87 ±1.71 / 204.17 ms │     no change │
│ QQuery 60 │        147.92 / 149.50 ±2.00 / 153.37 ms │        143.23 / 145.45 ±1.56 / 148.08 ms │     no change │
│ QQuery 61 │           13.64 / 13.76 ±0.10 / 13.90 ms │           13.25 / 13.72 ±0.53 / 14.68 ms │     no change │
│ QQuery 62 │      940.27 / 984.73 ±39.06 / 1042.39 ms │      905.86 / 957.44 ±45.11 / 1017.59 ms │     no change │
│ QQuery 63 │        105.63 / 107.46 ±0.96 / 108.28 ms │        102.93 / 105.77 ±1.79 / 108.03 ms │     no change │
│ QQuery 64 │        700.92 / 712.20 ±7.94 / 720.31 ms │        693.96 / 699.92 ±4.29 / 707.15 ms │     no change │
│ QQuery 65 │        266.45 / 271.37 ±4.05 / 277.86 ms │        255.86 / 259.20 ±3.55 / 265.78 ms │     no change │
│ QQuery 66 │       237.45 / 264.91 ±22.19 / 299.33 ms │       259.63 / 272.60 ±12.09 / 290.05 ms │     no change │
│ QQuery 67 │        331.19 / 339.16 ±6.54 / 350.67 ms │        315.33 / 322.81 ±8.18 / 338.08 ms │     no change │
│ QQuery 68 │            9.81 / 11.33 ±1.09 / 12.61 ms │             8.72 / 9.96 ±0.89 / 10.79 ms │ +1.14x faster │
│ QQuery 69 │        105.04 / 106.80 ±2.21 / 111.13 ms │        104.69 / 107.89 ±2.87 / 112.38 ms │     no change │
│ QQuery 70 │        352.69 / 360.69 ±8.14 / 376.03 ms │        347.96 / 359.69 ±8.31 / 369.89 ms │     no change │
│ QQuery 71 │        137.11 / 139.60 ±1.41 / 141.04 ms │        133.45 / 137.51 ±3.50 / 143.81 ms │     no change │
│ QQuery 72 │        640.13 / 651.43 ±7.70 / 662.45 ms │        621.92 / 632.04 ±6.08 / 640.61 ms │     no change │
│ QQuery 73 │              6.94 / 8.02 ±0.62 / 8.76 ms │              6.65 / 8.08 ±0.97 / 9.66 ms │     no change │
│ QQuery 74 │        652.50 / 659.01 ±5.33 / 666.47 ms │        589.57 / 599.77 ±8.17 / 608.00 ms │ +1.10x faster │
│ QQuery 75 │        280.28 / 282.78 ±2.14 / 285.74 ms │        274.85 / 279.23 ±2.73 / 282.82 ms │     no change │
│ QQuery 76 │        132.49 / 134.85 ±1.40 / 136.09 ms │        131.08 / 133.23 ±1.22 / 134.40 ms │     no change │
│ QQuery 77 │        191.93 / 193.74 ±1.98 / 197.44 ms │        188.33 / 190.21 ±1.43 / 192.58 ms │     no change │
│ QQuery 78 │        348.32 / 354.92 ±3.88 / 359.94 ms │        345.49 / 347.31 ±1.94 / 350.26 ms │     no change │
│ QQuery 79 │        248.47 / 251.51 ±2.34 / 254.93 ms │        232.22 / 236.53 ±3.05 / 240.14 ms │ +1.06x faster │
│ QQuery 80 │        325.22 / 327.42 ±2.18 / 331.43 ms │        320.26 / 321.97 ±0.97 / 322.85 ms │     no change │
│ QQuery 81 │           27.15 / 28.12 ±0.78 / 29.49 ms │           27.10 / 28.05 ±0.60 / 28.92 ms │     no change │
│ QQuery 82 │        203.56 / 206.66 ±3.09 / 212.06 ms │        201.11 / 204.23 ±1.65 / 205.72 ms │     no change │
│ QQuery 83 │           39.64 / 40.48 ±0.88 / 42.18 ms │           38.83 / 40.49 ±1.53 / 42.84 ms │     no change │
│ QQuery 84 │           49.67 / 51.24 ±1.41 / 53.39 ms │           48.03 / 48.91 ±0.79 / 49.99 ms │     no change │
│ QQuery 85 │        150.57 / 151.68 ±1.47 / 154.56 ms │        148.48 / 150.75 ±1.94 / 153.15 ms │     no change │
│ QQuery 86 │           41.09 / 41.78 ±0.58 / 42.53 ms │           39.61 / 40.55 ±0.68 / 41.41 ms │     no change │
│ QQuery 87 │           89.21 / 93.58 ±2.43 / 96.06 ms │           86.65 / 90.29 ±2.59 / 94.58 ms │     no change │
│ QQuery 88 │        101.78 / 103.59 ±1.03 / 104.76 ms │         99.53 / 100.67 ±0.90 / 102.29 ms │     no change │
│ QQuery 89 │        122.15 / 123.03 ±0.81 / 124.33 ms │        119.84 / 121.54 ±1.15 / 123.35 ms │     no change │
│ QQuery 90 │           25.14 / 26.35 ±0.73 / 27.11 ms │           22.82 / 23.84 ±0.73 / 24.84 ms │ +1.11x faster │
│ QQuery 91 │           64.48 / 65.72 ±0.98 / 67.18 ms │           61.82 / 64.28 ±1.45 / 65.76 ms │     no change │
│ QQuery 92 │           59.00 / 60.65 ±1.18 / 62.04 ms │           57.74 / 58.75 ±0.79 / 59.94 ms │     no change │
│ QQuery 93 │        190.44 / 194.41 ±2.46 / 198.07 ms │        188.60 / 189.82 ±1.22 / 191.43 ms │     no change │
│ QQuery 94 │           62.51 / 63.41 ±0.58 / 64.13 ms │           61.41 / 62.72 ±1.05 / 64.42 ms │     no change │
│ QQuery 95 │        130.21 / 132.23 ±1.13 / 133.61 ms │        128.85 / 129.54 ±0.99 / 131.46 ms │     no change │
│ QQuery 96 │           72.88 / 76.59 ±1.87 / 78.05 ms │           73.71 / 74.76 ±0.71 / 75.59 ms │     no change │
│ QQuery 97 │        128.75 / 131.93 ±1.76 / 133.96 ms │        124.44 / 126.16 ±0.91 / 127.13 ms │     no change │
│ QQuery 98 │        159.72 / 161.16 ±1.81 / 164.64 ms │        152.84 / 156.26 ±1.95 / 158.62 ms │     no change │
│ QQuery 99 │ 10913.00 / 10971.47 ±34.77 / 11015.63 ms │ 10807.53 / 10842.71 ±28.75 / 10892.11 ms │     no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                  │ 32690.68ms │
│ Total Time (alamb_reschedule_io)   │ 32076.06ms │
│ Average Time (HEAD)                │   330.21ms │
│ Average Time (alamb_reschedule_io) │   324.00ms │
│ Queries Faster                     │          9 │
│ Queries Slower                     │          1 │
│ Queries with No Change             │         89 │
│ Queries with Failure               │          0 │
└────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	163.8s
Peak memory	5.2 GiB
Avg memory	4.4 GiB
CPU user	270.2s
CPU sys	18.3s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	160.7s
Peak memory	5.4 GiB
Avg memory	4.3 GiB
CPU user	265.2s
CPU sys	17.4s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

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

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 alamb_reschedule_io
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                       HEAD ┃                       alamb_reschedule_io ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │                6.27 / 6.68 ±0.71 / 8.10 ms │               6.53 / 6.98 ±0.72 / 8.41 ms │     no change │
│ QQuery 2  │         113.07 / 133.94 ±24.30 / 165.56 ms │        112.53 / 122.96 ±18.73 / 160.37 ms │ +1.09x faster │
│ QQuery 3  │          109.98 / 112.34 ±1.85 / 115.47 ms │         109.19 / 110.48 ±0.89 / 111.63 ms │     no change │
│ QQuery 4  │      1072.70 / 1093.08 ±11.31 / 1105.64 ms │     1059.62 / 1088.52 ±15.29 / 1101.76 ms │     no change │
│ QQuery 5  │          188.82 / 195.81 ±4.34 / 202.39 ms │         189.38 / 193.26 ±2.74 / 197.00 ms │     no change │
│ QQuery 6  │          252.78 / 261.81 ±6.65 / 269.60 ms │         253.18 / 259.74 ±4.53 / 267.13 ms │     no change │
│ QQuery 7  │          334.32 / 341.14 ±5.22 / 349.19 ms │         331.88 / 335.86 ±2.89 / 339.58 ms │     no change │
│ QQuery 8  │          153.09 / 161.36 ±4.25 / 164.98 ms │         155.38 / 160.79 ±4.40 / 166.46 ms │     no change │
│ QQuery 9  │         218.84 / 244.07 ±19.91 / 268.70 ms │        222.68 / 246.47 ±20.85 / 274.12 ms │     no change │
│ QQuery 10 │          169.81 / 176.08 ±4.15 / 182.35 ms │         169.79 / 176.23 ±5.48 / 185.97 ms │     no change │
│ QQuery 11 │          712.00 / 716.25 ±3.59 / 720.58 ms │        685.39 / 711.52 ±14.65 / 727.80 ms │     no change │
│ QQuery 12 │             37.04 / 39.32 ±1.51 / 41.77 ms │            38.13 / 41.01 ±2.52 / 44.98 ms │     no change │
│ QQuery 13 │          564.29 / 570.11 ±5.31 / 579.91 ms │         558.34 / 567.16 ±8.74 / 582.46 ms │     no change │
│ QQuery 14 │          900.62 / 915.54 ±8.33 / 922.67 ms │         914.47 / 921.81 ±4.93 / 928.69 ms │     no change │
│ QQuery 15 │             19.16 / 21.50 ±2.21 / 25.62 ms │            19.55 / 22.24 ±2.87 / 27.65 ms │     no change │
│ QQuery 16 │                6.86 / 7.45 ±0.50 / 8.14 ms │               6.87 / 7.12 ±0.20 / 7.34 ms │     no change │
│ QQuery 17 │          199.69 / 206.42 ±3.81 / 210.56 ms │         200.83 / 203.87 ±2.22 / 206.99 ms │     no change │
│ QQuery 18 │         425.56 / 470.09 ±24.26 / 494.72 ms │        412.23 / 447.05 ±26.63 / 479.18 ms │     no change │
│ QQuery 19 │          142.14 / 145.74 ±2.64 / 149.80 ms │         141.10 / 144.11 ±2.45 / 147.16 ms │     no change │
│ QQuery 20 │             15.07 / 17.35 ±2.23 / 21.27 ms │            15.99 / 17.04 ±1.41 / 19.64 ms │     no change │
│ QQuery 21 │             27.12 / 28.35 ±0.74 / 29.35 ms │            26.82 / 28.09 ±0.75 / 29.15 ms │     no change │
│ QQuery 22 │          513.34 / 520.77 ±5.82 / 529.21 ms │         499.84 / 509.60 ±7.99 / 523.82 ms │     no change │
│ QQuery 23 │      1207.87 / 1242.56 ±28.41 / 1292.55 ms │      1224.14 / 1231.57 ±8.13 / 1244.72 ms │     no change │
│ QQuery 24 │          120.38 / 125.98 ±3.11 / 129.86 ms │         123.61 / 126.37 ±1.53 / 128.32 ms │     no change │
│ QQuery 25 │          294.63 / 302.36 ±4.50 / 307.97 ms │         299.79 / 304.44 ±3.40 / 308.64 ms │     no change │
│ QQuery 26 │         143.20 / 154.39 ±10.55 / 167.96 ms │         145.72 / 152.60 ±7.33 / 166.06 ms │     no change │
│ QQuery 27 │                6.49 / 6.67 ±0.14 / 6.88 ms │               6.62 / 7.42 ±0.61 / 8.27 ms │  1.11x slower │
│ QQuery 28 │          227.27 / 231.64 ±2.54 / 234.18 ms │         220.49 / 231.84 ±7.34 / 240.82 ms │     no change │
│ QQuery 29 │          252.93 / 256.28 ±3.02 / 261.76 ms │         234.18 / 249.71 ±8.33 / 256.93 ms │     no change │
│ QQuery 30 │             55.73 / 58.75 ±1.65 / 60.61 ms │            58.04 / 60.33 ±2.21 / 64.25 ms │     no change │
│ QQuery 31 │          176.93 / 179.61 ±2.61 / 184.52 ms │         174.87 / 176.71 ±2.69 / 182.03 ms │     no change │
│ QQuery 32 │             52.73 / 54.80 ±1.71 / 57.71 ms │            52.89 / 53.91 ±0.94 / 55.55 ms │     no change │
│ QQuery 33 │          133.40 / 135.22 ±2.40 / 139.87 ms │         131.02 / 133.37 ±1.36 / 134.75 ms │     no change │
│ QQuery 34 │                7.19 / 7.86 ±0.77 / 9.35 ms │               6.88 / 7.43 ±0.60 / 8.43 ms │ +1.06x faster │
│ QQuery 35 │          148.37 / 152.18 ±2.52 / 155.19 ms │         144.14 / 150.69 ±3.52 / 154.41 ms │     no change │
│ QQuery 36 │                6.55 / 6.84 ±0.39 / 7.61 ms │               6.36 / 6.56 ±0.23 / 6.92 ms │     no change │
│ QQuery 37 │                4.34 / 4.47 ±0.09 / 4.58 ms │               4.32 / 4.57 ±0.24 / 4.90 ms │     no change │
│ QQuery 38 │           99.46 / 107.24 ±5.34 / 115.93 ms │         106.29 / 108.72 ±1.84 / 111.40 ms │     no change │
│ QQuery 39 │          137.40 / 138.82 ±1.63 / 141.92 ms │         136.14 / 139.22 ±3.25 / 145.24 ms │     no change │
│ QQuery 40 │          134.60 / 141.20 ±6.78 / 153.55 ms │         134.09 / 139.47 ±6.55 / 151.74 ms │     no change │
│ QQuery 41 │             14.01 / 14.92 ±0.65 / 15.94 ms │            14.22 / 14.98 ±0.67 / 15.79 ms │     no change │
│ QQuery 42 │          113.57 / 116.28 ±1.69 / 118.06 ms │         112.79 / 116.15 ±2.11 / 118.98 ms │     no change │
│ QQuery 43 │                5.74 / 6.11 ±0.47 / 6.98 ms │               5.94 / 6.07 ±0.18 / 6.42 ms │     no change │
│ QQuery 44 │             11.29 / 11.92 ±0.79 / 13.43 ms │            11.39 / 12.40 ±0.70 / 13.29 ms │     no change │
│ QQuery 45 │             41.76 / 44.41 ±1.83 / 47.26 ms │            43.62 / 46.78 ±2.10 / 50.23 ms │  1.05x slower │
│ QQuery 46 │               8.18 / 8.90 ±0.74 / 10.26 ms │               8.51 / 8.95 ±0.44 / 9.74 ms │     no change │
│ QQuery 47 │         773.69 / 792.14 ±10.88 / 803.86 ms │        773.57 / 785.18 ±10.47 / 797.25 ms │     no change │
│ QQuery 48 │          483.79 / 499.26 ±9.56 / 510.87 ms │         484.30 / 493.24 ±6.64 / 504.41 ms │     no change │
│ QQuery 49 │          282.72 / 289.17 ±3.37 / 291.65 ms │         283.79 / 285.95 ±1.94 / 288.62 ms │     no change │
│ QQuery 50 │          537.60 / 541.78 ±5.35 / 552.18 ms │         516.91 / 535.02 ±9.37 / 544.21 ms │     no change │
│ QQuery 51 │          216.96 / 224.02 ±4.51 / 230.43 ms │         217.12 / 220.25 ±2.45 / 223.52 ms │     no change │
│ QQuery 52 │          112.88 / 116.49 ±2.55 / 119.33 ms │         107.79 / 114.40 ±3.65 / 118.35 ms │     no change │
│ QQuery 53 │          138.56 / 145.71 ±5.72 / 152.09 ms │         139.01 / 146.57 ±4.80 / 152.46 ms │     no change │
│ QQuery 54 │          128.99 / 133.25 ±3.78 / 139.10 ms │         127.28 / 130.24 ±1.62 / 132.23 ms │     no change │
│ QQuery 55 │          115.50 / 116.37 ±0.79 / 117.79 ms │         111.84 / 115.34 ±2.20 / 118.80 ms │     no change │
│ QQuery 56 │          134.75 / 137.57 ±1.95 / 139.88 ms │         132.49 / 133.49 ±0.71 / 134.37 ms │     no change │
│ QQuery 57 │          190.10 / 192.81 ±1.92 / 195.60 ms │         189.96 / 191.96 ±2.06 / 195.51 ms │     no change │
│ QQuery 58 │          372.72 / 376.91 ±2.92 / 381.19 ms │         369.03 / 373.39 ±3.48 / 379.25 ms │     no change │
│ QQuery 59 │          277.86 / 283.17 ±3.12 / 286.04 ms │         276.74 / 283.34 ±4.36 / 287.35 ms │     no change │
│ QQuery 60 │          137.07 / 140.68 ±2.08 / 143.21 ms │         137.78 / 139.71 ±1.65 / 142.54 ms │     no change │
│ QQuery 61 │             12.36 / 12.98 ±0.60 / 14.04 ms │            12.18 / 12.80 ±0.65 / 13.81 ms │     no change │
│ QQuery 62 │         910.03 / 932.26 ±16.07 / 958.39 ms │       893.57 / 954.07 ±57.21 / 1035.00 ms │     no change │
│ QQuery 63 │          140.01 / 148.44 ±5.49 / 154.54 ms │         138.53 / 146.66 ±8.08 / 159.42 ms │     no change │
│ QQuery 64 │ 28789.27 / 29942.80 ±1155.86 / 31929.09 ms │ 27924.72 / 28745.62 ±672.55 / 29932.13 ms │     no change │
│ QQuery 65 │          364.30 / 372.40 ±4.69 / 378.79 ms │         357.50 / 370.35 ±9.53 / 384.69 ms │     no change │
│ QQuery 66 │          196.86 / 199.68 ±2.68 / 204.67 ms │         191.15 / 199.45 ±4.95 / 206.01 ms │     no change │
│ QQuery 67 │         495.18 / 508.83 ±11.81 / 530.81 ms │         504.25 / 510.36 ±5.15 / 516.95 ms │     no change │
│ QQuery 68 │              8.43 / 10.92 ±2.94 / 16.01 ms │             9.74 / 11.32 ±1.31 / 13.05 ms │     no change │
│ QQuery 69 │          167.10 / 174.80 ±7.92 / 187.81 ms │         170.05 / 179.58 ±6.46 / 190.19 ms │     no change │
│ QQuery 70 │          428.32 / 440.77 ±7.50 / 450.37 ms │         426.58 / 437.82 ±8.72 / 449.47 ms │     no change │
│ QQuery 71 │          129.77 / 133.62 ±2.45 / 136.12 ms │         128.55 / 130.71 ±1.56 / 132.96 ms │     no change │
│ QQuery 72 │       1198.52 / 1209.29 ±6.06 / 1216.07 ms │      1194.74 / 1202.25 ±4.90 / 1209.20 ms │     no change │
│ QQuery 73 │                6.87 / 6.97 ±0.11 / 7.14 ms │               6.45 / 7.01 ±0.59 / 8.08 ms │     no change │
│ QQuery 74 │          506.53 / 514.37 ±4.47 / 519.77 ms │         486.98 / 500.57 ±8.09 / 508.66 ms │     no change │
│ QQuery 75 │          285.44 / 292.02 ±3.81 / 296.92 ms │         280.82 / 288.35 ±4.58 / 295.26 ms │     no change │
│ QQuery 76 │          277.85 / 282.33 ±3.05 / 287.21 ms │         271.40 / 279.45 ±6.56 / 291.27 ms │     no change │
│ QQuery 77 │          226.77 / 230.66 ±2.56 / 234.84 ms │         227.32 / 229.42 ±1.60 / 231.54 ms │     no change │
│ QQuery 78 │          305.85 / 311.60 ±4.89 / 317.46 ms │         303.53 / 309.21 ±4.31 / 314.67 ms │     no change │
│ QQuery 79 │          261.44 / 268.24 ±3.50 / 271.41 ms │         257.72 / 265.76 ±4.54 / 270.43 ms │     no change │
│ QQuery 80 │          324.42 / 327.98 ±2.28 / 331.01 ms │         322.12 / 327.21 ±3.69 / 331.70 ms │     no change │
│ QQuery 81 │             30.05 / 33.94 ±3.19 / 38.17 ms │            30.84 / 32.90 ±1.85 / 35.63 ms │     no change │
│ QQuery 82 │          188.47 / 195.20 ±3.68 / 198.51 ms │         188.20 / 194.12 ±3.02 / 196.63 ms │     no change │
│ QQuery 83 │             46.34 / 47.50 ±1.12 / 49.44 ms │            44.65 / 45.79 ±0.82 / 47.12 ms │     no change │
│ QQuery 84 │             64.23 / 66.54 ±1.27 / 67.75 ms │            64.93 / 66.14 ±1.03 / 67.99 ms │     no change │
│ QQuery 85 │          273.58 / 283.57 ±5.22 / 288.84 ms │         281.64 / 288.05 ±4.61 / 293.63 ms │     no change │
│ QQuery 86 │             47.77 / 49.48 ±1.53 / 52.13 ms │            48.16 / 49.25 ±0.82 / 50.16 ms │     no change │
│ QQuery 87 │          109.47 / 112.23 ±1.59 / 114.13 ms │         107.71 / 111.03 ±3.37 / 115.41 ms │     no change │
│ QQuery 88 │          128.64 / 132.90 ±3.41 / 138.30 ms │         130.54 / 134.23 ±2.33 / 137.53 ms │     no change │
│ QQuery 89 │          158.85 / 162.31 ±2.86 / 166.96 ms │         151.66 / 161.89 ±7.39 / 170.11 ms │     no change │
│ QQuery 90 │             25.62 / 26.30 ±0.47 / 26.96 ms │            26.61 / 29.54 ±2.02 / 31.95 ms │  1.12x slower │
│ QQuery 91 │          100.12 / 104.53 ±3.90 / 109.95 ms │          97.72 / 107.17 ±6.19 / 116.92 ms │     no change │
│ QQuery 92 │             54.70 / 56.54 ±1.60 / 58.72 ms │            53.75 / 54.34 ±0.54 / 55.13 ms │     no change │
│ QQuery 93 │          189.01 / 191.09 ±1.70 / 192.72 ms │         189.04 / 191.86 ±2.21 / 195.74 ms │     no change │
│ QQuery 94 │             63.88 / 65.66 ±1.16 / 67.02 ms │            63.41 / 64.57 ±0.87 / 66.07 ms │     no change │
│ QQuery 95 │          146.47 / 146.86 ±0.29 / 147.13 ms │         143.36 / 145.28 ±1.59 / 147.74 ms │     no change │
│ QQuery 96 │             87.46 / 88.88 ±1.03 / 90.49 ms │            79.93 / 86.67 ±3.39 / 88.70 ms │     no change │
│ QQuery 97 │          153.35 / 156.67 ±2.24 / 159.64 ms │         154.31 / 156.59 ±1.43 / 158.32 ms │     no change │
│ QQuery 98 │          122.05 / 124.97 ±1.65 / 126.76 ms │         118.20 / 123.66 ±3.72 / 128.14 ms │     no change │
│ QQuery 99 │   10797.56 / 10812.70 ±14.89 / 10837.67 ms │  10752.72 / 10779.81 ±21.36 / 10810.05 ms │     no change │
└───────────┴────────────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                  ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                  │ 63062.79ms │
│ Total Time (alamb_reschedule_io)   │ 61691.12ms │
│ Average Time (HEAD)                │   637.00ms │
│ Average Time (alamb_reschedule_io) │   623.14ms │
│ Queries Faster                     │          2 │
│ Queries Slower                     │          3 │
│ Queries with No Change             │         94 │
│ Queries with Failure               │          0 │
└────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	315.7s
Peak memory	39.8 GiB
Avg memory	27.9 GiB
CPU user	625.2s
CPU sys	28.5s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	308.9s
Peak memory	35.6 GiB
Avg memory	25.3 GiB
CPU user	606.6s
CPU sys	27.6s
Peak spill	0 B

---

File an issue against this benchmark runner

[Dandandan]

│ Total Time (HEAD)                  │ 20149.48ms │
│ Total Time (alamb_reschedule_io)   │ 17963.47ms │

(With filter pushdown, Clicbench is under 18s!)

[adriangb]

Let's ship it 📦 🚀 !!

[comphead]

Looking fwd it!

[comphead]

Thanks @alamb