Skip to content

fix: handle interleaved HashJoin projections in sort pushdown#23591

Open
xudong963 wants to merge 1 commit into
apache:mainfrom
massive-com:xudong963/fix-interleaved-hash-join-sort-pushdown-upstream
Open

fix: handle interleaved HashJoin projections in sort pushdown#23591
xudong963 wants to merge 1 commit into
apache:mainfrom
massive-com:xudong963/fix-interleaved-hash-join-sort-pushdown-upstream

Conversation

@xudong963

Copy link
Copy Markdown
Member

Which issue does this PR close?

Rationale for this change

Sort pushdown through a HashJoinExec with an embedded projection assumed that all projected left-side columns formed a contiguous prefix. That assumption is not valid when the projection interleaves columns from the two join inputs.

For a CollectLeft right join projected as [right.col_a, left.nullable_col], a required sort on output column 1 was therefore mistaken for a sort from the right child. The optimizer pushed a sort on right.col_a below the join, and SanityCheckPlan rejected the resulting plan because it did not satisfy the parent sort on left.nullable_col.

What changes are included in this PR?

  • Determine whether required sort columns come from the right child by inspecting each projected output column's JoinSide, instead of comparing output positions with the number of projected left columns.
  • Add a regression test with an interleaved hash join projection. The test runs SanityCheckPlan and snapshots the valid optimized plan.

Are these changes tested?

Yes.

  • On unmodified main, the new regression test fails in SanityCheckPlan: the right input is sorted by col_a, while the parent requires nullable_col.
  • With this change, the focused regression test passes.
  • cargo fmt --all
  • cargo clippy --all-targets --all-features -- -D warnings
  • RUST_BACKTRACE=1 cargo test --profile ci --exclude datafusion-examples --exclude datafusion-benchmarks --exclude datafusion-cli --workspace --lib --tests --bins --features avro,json,backtrace,extended_tests,recursive_protection,parquet_encryption

Are there any user-facing changes?

There are no API changes. Queries whose hash join projection interleaves columns from both inputs no longer receive an invalid sort pushdown that can fail physical-plan sanity checking.

@github-actions github-actions Bot added optimizer Optimizer rules core Core DataFusion crate labels Jul 15, 2026
@zhuqi-lucas

Copy link
Copy Markdown
Contributor

run benchmarks

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4987822131-1089-m2rh7 6.12.85+ #1 SMP Mon May 11 08:17:35 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 xudong963/fix-interleaved-hash-join-sort-pushdown-upstream (ad02f44) to 4b68de6 (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4987822131-1090-55wf4 6.12.85+ #1 SMP Mon May 11 08:17:35 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 xudong963/fix-interleaved-hash-join-sort-pushdown-upstream (ad02f44) to 4b68de6 (merge-base) diff using: tpcds
Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4987822131-1091-8qfwd 6.12.85+ #1 SMP Mon May 11 08:17:35 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 xudong963/fix-interleaved-hash-join-sort-pushdown-upstream (ad02f44) to 4b68de6 (merge-base) diff using: tpch
Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 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 xudong963_fix-interleaved-hash-join-sort-pushdown-upstream
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ xudong963_fix-interleaved-hash-join-sort-pushdown-upstream ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │ 38.48 / 39.29 ±1.04 / 41.22 ms │                             38.11 / 39.42 ±1.43 / 41.29 ms │    no change │
│ QQuery 2  │ 19.32 / 19.48 ±0.13 / 19.69 ms │                             18.87 / 19.24 ±0.27 / 19.64 ms │    no change │
│ QQuery 3  │ 33.00 / 34.05 ±1.47 / 36.97 ms │                             31.89 / 33.43 ±1.20 / 35.56 ms │    no change │
│ QQuery 4  │ 17.14 / 17.82 ±0.86 / 19.50 ms │                             17.25 / 17.49 ±0.15 / 17.71 ms │    no change │
│ QQuery 5  │ 37.31 / 38.39 ±0.97 / 39.68 ms │                             39.60 / 40.63 ±1.15 / 42.79 ms │ 1.06x slower │
│ QQuery 6  │ 16.13 / 16.85 ±1.30 / 19.44 ms │                             15.98 / 16.83 ±0.87 / 18.49 ms │    no change │
│ QQuery 7  │ 43.94 / 45.55 ±1.23 / 46.67 ms │                             43.01 / 44.55 ±1.49 / 47.27 ms │    no change │
│ QQuery 8  │ 42.55 / 42.86 ±0.28 / 43.33 ms │                             42.51 / 43.20 ±1.02 / 45.22 ms │    no change │
│ QQuery 9  │ 49.31 / 50.47 ±0.97 / 51.83 ms │                             49.50 / 51.05 ±1.23 / 52.86 ms │    no change │
│ QQuery 10 │ 41.96 / 42.25 ±0.28 / 42.77 ms │                             41.94 / 42.55 ±0.72 / 43.96 ms │    no change │
│ QQuery 11 │ 13.44 / 14.02 ±0.87 / 15.75 ms │                             13.26 / 13.45 ±0.14 / 13.59 ms │    no change │
│ QQuery 12 │ 24.02 / 24.60 ±0.53 / 25.45 ms │                             24.05 / 24.23 ±0.24 / 24.69 ms │    no change │
│ QQuery 13 │ 31.59 / 34.02 ±2.51 / 38.48 ms │                             31.80 / 33.15 ±1.98 / 37.07 ms │    no change │
│ QQuery 14 │ 23.50 / 23.67 ±0.13 / 23.83 ms │                             23.30 / 23.55 ±0.25 / 23.94 ms │    no change │
│ QQuery 15 │ 31.13 / 32.08 ±0.75 / 32.75 ms │                             30.79 / 31.33 ±0.73 / 32.76 ms │    no change │
│ QQuery 16 │ 14.13 / 14.21 ±0.08 / 14.35 ms │                             13.94 / 14.12 ±0.16 / 14.34 ms │    no change │
│ QQuery 17 │ 70.41 / 71.61 ±1.06 / 73.43 ms │                             70.73 / 71.49 ±0.86 / 73.14 ms │    no change │
│ QQuery 18 │ 58.57 / 59.78 ±1.13 / 61.75 ms │                             58.33 / 59.36 ±1.18 / 61.58 ms │    no change │
│ QQuery 19 │ 33.02 / 33.82 ±1.38 / 36.57 ms │                             32.78 / 33.23 ±0.37 / 33.75 ms │    no change │
│ QQuery 20 │ 31.93 / 32.15 ±0.28 / 32.68 ms │                             31.50 / 32.87 ±1.44 / 35.43 ms │    no change │
│ QQuery 21 │ 54.34 / 56.07 ±1.12 / 57.30 ms │                             56.34 / 57.36 ±0.88 / 58.81 ms │    no change │
│ QQuery 22 │ 13.52 / 13.80 ±0.19 / 14.12 ms │                             13.61 / 13.83 ±0.17 / 14.05 ms │    no change │
└───────────┴────────────────────────────────┴────────────────────────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                                         ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                                         │ 756.84ms │
│ Total Time (xudong963_fix-interleaved-hash-join-sort-pushdown-upstream)   │ 756.38ms │
│ Average Time (HEAD)                                                       │  34.40ms │
│ Average Time (xudong963_fix-interleaved-hash-join-sort-pushdown-upstream) │  34.38ms │
│ Queries Faster                                                            │        0 │
│ Queries Slower                                                            │        1 │
│ Queries with No Change                                                    │       21 │
│ Queries with Failure                                                      │        0 │
└───────────────────────────────────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.3 GiB
Avg memory 535.1 MiB
CPU user 21.5s
CPU sys 1.7s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.3 GiB
Avg memory 531.8 MiB
CPU user 21.7s
CPU sys 1.6s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 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 xudong963_fix-interleaved-hash-join-sort-pushdown-upstream
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ xudong963_fix-interleaved-hash-join-sort-pushdown-upstream ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.37 / 5.94 ±0.95 / 7.84 ms │                                5.68 / 6.17 ±0.84 / 7.85 ms │     no change │
│ QQuery 2  │        79.92 / 80.39 ±0.32 / 80.86 ms │                             80.76 / 80.98 ±0.19 / 81.23 ms │     no change │
│ QQuery 3  │        29.80 / 29.89 ±0.09 / 30.06 ms │                             29.64 / 29.93 ±0.26 / 30.32 ms │     no change │
│ QQuery 4  │     517.26 / 521.12 ±3.20 / 524.89 ms │                          509.88 / 516.77 ±4.96 / 523.13 ms │     no change │
│ QQuery 5  │        51.97 / 52.07 ±0.08 / 52.18 ms │                             52.24 / 52.60 ±0.33 / 53.06 ms │     no change │
│ QQuery 6  │        37.43 / 37.61 ±0.15 / 37.81 ms │                             37.04 / 37.52 ±0.39 / 38.02 ms │     no change │
│ QQuery 7  │        95.45 / 96.30 ±1.30 / 98.88 ms │                            94.59 / 96.49 ±2.78 / 102.00 ms │     no change │
│ QQuery 8  │        37.12 / 37.48 ±0.38 / 38.14 ms │                             37.40 / 37.72 ±0.34 / 38.38 ms │     no change │
│ QQuery 9  │        53.75 / 55.26 ±1.68 / 58.35 ms │                             51.76 / 55.23 ±1.91 / 57.07 ms │     no change │
│ QQuery 10 │        63.68 / 64.12 ±0.33 / 64.52 ms │                             64.16 / 64.89 ±0.44 / 65.41 ms │     no change │
│ QQuery 11 │     320.29 / 324.50 ±2.87 / 328.05 ms │                          320.99 / 322.52 ±1.23 / 324.61 ms │     no change │
│ QQuery 12 │        29.21 / 29.37 ±0.23 / 29.84 ms │                             29.21 / 29.58 ±0.39 / 30.27 ms │     no change │
│ QQuery 13 │     119.17 / 121.24 ±2.30 / 125.68 ms │                          120.26 / 121.35 ±1.08 / 123.24 ms │     no change │
│ QQuery 14 │     416.07 / 420.56 ±5.15 / 430.53 ms │                          416.27 / 420.83 ±3.19 / 425.42 ms │     no change │
│ QQuery 15 │        59.47 / 59.93 ±0.29 / 60.25 ms │                             59.37 / 60.25 ±0.70 / 61.29 ms │     no change │
│ QQuery 16 │           6.72 / 6.96 ±0.17 / 7.19 ms │                                6.53 / 6.73 ±0.26 / 7.23 ms │     no change │
│ QQuery 17 │        81.80 / 83.03 ±1.23 / 84.76 ms │                             80.20 / 81.79 ±1.39 / 83.71 ms │     no change │
│ QQuery 18 │     124.42 / 125.44 ±0.85 / 126.83 ms │                          125.14 / 125.96 ±0.84 / 127.47 ms │     no change │
│ QQuery 19 │        42.23 / 42.88 ±0.72 / 44.23 ms │                             42.26 / 42.45 ±0.18 / 42.73 ms │     no change │
│ QQuery 20 │        36.58 / 37.36 ±1.02 / 39.34 ms │                             36.43 / 38.24 ±1.36 / 40.50 ms │     no change │
│ QQuery 21 │        17.85 / 18.14 ±0.26 / 18.49 ms │                             18.16 / 18.31 ±0.22 / 18.75 ms │     no change │
│ QQuery 22 │        63.89 / 64.94 ±1.05 / 66.56 ms │                             63.78 / 64.91 ±0.94 / 66.49 ms │     no change │
│ QQuery 23 │     352.01 / 356.59 ±3.60 / 360.92 ms │                          355.65 / 359.42 ±2.62 / 363.41 ms │     no change │
│ QQuery 24 │     226.32 / 229.61 ±4.31 / 238.05 ms │                          227.28 / 230.50 ±4.23 / 238.49 ms │     no change │
│ QQuery 25 │     111.15 / 114.54 ±3.95 / 122.20 ms │                          110.98 / 114.44 ±5.07 / 124.45 ms │     no change │
│ QQuery 26 │        57.86 / 58.89 ±0.75 / 59.84 ms │                             58.14 / 58.84 ±0.36 / 59.11 ms │     no change │
│ QQuery 27 │           6.17 / 6.37 ±0.21 / 6.78 ms │                                6.33 / 6.47 ±0.12 / 6.69 ms │     no change │
│ QQuery 28 │        60.92 / 61.43 ±0.36 / 61.83 ms │                             57.58 / 60.07 ±1.81 / 62.15 ms │     no change │
│ QQuery 29 │      98.12 / 100.86 ±2.65 / 104.70 ms │                           98.36 / 101.38 ±2.96 / 105.12 ms │     no change │
│ QQuery 30 │        33.44 / 33.95 ±0.53 / 34.85 ms │                             33.51 / 33.79 ±0.25 / 34.20 ms │     no change │
│ QQuery 31 │     112.67 / 112.97 ±0.24 / 113.26 ms │                          112.51 / 112.87 ±0.41 / 113.66 ms │     no change │
│ QQuery 32 │        21.35 / 23.14 ±2.51 / 28.13 ms │                             20.92 / 22.64 ±3.27 / 29.17 ms │     no change │
│ QQuery 33 │        38.45 / 39.58 ±0.93 / 41.26 ms │                             38.67 / 39.12 ±0.27 / 39.44 ms │     no change │
│ QQuery 34 │        10.15 / 10.23 ±0.05 / 10.28 ms │                             10.10 / 10.58 ±0.36 / 11.18 ms │     no change │
│ QQuery 35 │        74.54 / 75.12 ±0.43 / 75.81 ms │                             74.82 / 75.75 ±0.48 / 76.11 ms │     no change │
│ QQuery 36 │           5.59 / 5.87 ±0.21 / 6.20 ms │                                5.89 / 6.01 ±0.15 / 6.28 ms │     no change │
│ QQuery 37 │           6.84 / 7.01 ±0.09 / 7.11 ms │                                6.74 / 7.00 ±0.13 / 7.12 ms │     no change │
│ QQuery 38 │        64.30 / 64.95 ±0.48 / 65.38 ms │                             64.21 / 65.39 ±1.37 / 67.98 ms │     no change │
│ QQuery 39 │        92.11 / 92.76 ±0.69 / 93.99 ms │                             92.06 / 92.74 ±0.65 / 93.86 ms │     no change │
│ QQuery 40 │        23.70 / 24.11 ±0.42 / 24.90 ms │                             23.72 / 23.95 ±0.15 / 24.18 ms │     no change │
│ QQuery 41 │        11.80 / 11.97 ±0.15 / 12.22 ms │                             11.67 / 11.85 ±0.12 / 12.03 ms │     no change │
│ QQuery 42 │        23.98 / 24.37 ±0.35 / 25.04 ms │                             24.45 / 24.92 ±0.33 / 25.29 ms │     no change │
│ QQuery 43 │           4.89 / 5.04 ±0.14 / 5.31 ms │                                4.97 / 5.11 ±0.18 / 5.44 ms │     no change │
│ QQuery 44 │           9.39 / 9.55 ±0.11 / 9.69 ms │                                9.60 / 9.69 ±0.10 / 9.83 ms │     no change │
│ QQuery 45 │        40.57 / 43.06 ±2.13 / 45.81 ms │                             40.84 / 42.76 ±2.69 / 48.07 ms │     no change │
│ QQuery 46 │        12.27 / 12.79 ±0.43 / 13.31 ms │                             12.68 / 13.02 ±0.36 / 13.62 ms │     no change │
│ QQuery 47 │     238.07 / 242.65 ±4.10 / 248.96 ms │                          238.54 / 244.02 ±5.39 / 254.01 ms │     no change │
│ QQuery 48 │        96.73 / 97.32 ±0.66 / 98.60 ms │                             96.37 / 96.81 ±0.52 / 97.50 ms │     no change │
│ QQuery 49 │        76.37 / 77.37 ±1.53 / 80.39 ms │                             76.89 / 77.61 ±0.72 / 78.74 ms │     no change │
│ QQuery 50 │        59.79 / 60.51 ±0.66 / 61.61 ms │                             59.57 / 59.98 ±0.30 / 60.48 ms │     no change │
│ QQuery 51 │        91.12 / 93.44 ±1.47 / 95.20 ms │                            92.11 / 95.34 ±4.32 / 103.81 ms │     no change │
│ QQuery 52 │        24.55 / 25.84 ±1.68 / 29.11 ms │                             24.67 / 25.32 ±0.44 / 26.00 ms │     no change │
│ QQuery 53 │        30.24 / 30.97 ±0.50 / 31.73 ms │                             30.15 / 30.40 ±0.20 / 30.69 ms │     no change │
│ QQuery 54 │        55.75 / 56.53 ±0.73 / 57.92 ms │                             56.06 / 56.36 ±0.27 / 56.78 ms │     no change │
│ QQuery 55 │        23.68 / 24.02 ±0.42 / 24.81 ms │                             23.69 / 23.92 ±0.21 / 24.18 ms │     no change │
│ QQuery 56 │        39.80 / 40.04 ±0.19 / 40.34 ms │                             39.19 / 39.70 ±0.37 / 40.21 ms │     no change │
│ QQuery 57 │     177.49 / 179.64 ±2.55 / 184.53 ms │                          177.56 / 179.33 ±1.43 / 181.18 ms │     no change │
│ QQuery 58 │     114.95 / 118.42 ±2.67 / 121.90 ms │                          116.02 / 118.31 ±1.77 / 120.21 ms │     no change │
│ QQuery 59 │     117.43 / 118.08 ±0.36 / 118.51 ms │                          118.00 / 118.43 ±0.30 / 118.72 ms │     no change │
│ QQuery 60 │        40.07 / 41.12 ±0.99 / 42.71 ms │                             40.17 / 42.45 ±1.97 / 45.85 ms │     no change │
│ QQuery 61 │        12.12 / 12.39 ±0.19 / 12.68 ms │                             12.42 / 12.52 ±0.13 / 12.78 ms │     no change │
│ QQuery 62 │        46.58 / 46.81 ±0.25 / 47.25 ms │                             46.59 / 47.17 ±0.32 / 47.54 ms │     no change │
│ QQuery 63 │        30.05 / 30.38 ±0.30 / 30.85 ms │                             29.83 / 30.01 ±0.14 / 30.24 ms │     no change │
│ QQuery 64 │     411.20 / 416.23 ±3.82 / 420.17 ms │                          414.35 / 418.55 ±3.24 / 423.36 ms │     no change │
│ QQuery 65 │     129.85 / 131.91 ±2.81 / 137.42 ms │                          126.75 / 132.11 ±4.27 / 139.86 ms │     no change │
│ QQuery 66 │        80.97 / 81.92 ±0.98 / 83.62 ms │                             80.73 / 81.02 ±0.38 / 81.73 ms │     no change │
│ QQuery 67 │     254.55 / 258.57 ±3.25 / 263.66 ms │                          258.25 / 263.01 ±4.49 / 268.93 ms │     no change │
│ QQuery 68 │        12.40 / 13.51 ±1.75 / 16.96 ms │                             12.20 / 12.39 ±0.12 / 12.53 ms │ +1.09x faster │
│ QQuery 69 │        58.10 / 58.50 ±0.46 / 59.22 ms │                             58.59 / 59.23 ±0.59 / 60.25 ms │     no change │
│ QQuery 70 │     106.38 / 108.83 ±2.02 / 111.94 ms │                          107.09 / 108.08 ±1.08 / 110.07 ms │     no change │
│ QQuery 71 │        36.12 / 38.67 ±3.02 / 44.17 ms │                             36.27 / 40.11 ±6.26 / 52.59 ms │     no change │
│ QQuery 72 │ 2022.86 / 2060.23 ±44.28 / 2133.30 ms │                      2149.81 / 2213.11 ±43.89 / 2281.87 ms │  1.07x slower │
│ QQuery 73 │         9.92 / 10.23 ±0.29 / 10.61 ms │                              9.85 / 10.20 ±0.25 / 10.48 ms │     no change │
│ QQuery 74 │     182.34 / 185.54 ±2.46 / 188.86 ms │                          182.35 / 187.36 ±5.66 / 196.88 ms │     no change │
│ QQuery 75 │     150.20 / 152.97 ±2.68 / 158.02 ms │                          149.69 / 150.90 ±1.31 / 153.39 ms │     no change │
│ QQuery 76 │        35.85 / 36.63 ±0.74 / 37.84 ms │                             36.32 / 38.38 ±2.31 / 42.78 ms │     no change │
│ QQuery 77 │        61.66 / 62.15 ±0.57 / 63.24 ms │                             62.01 / 62.59 ±0.41 / 63.27 ms │     no change │
│ QQuery 78 │     200.90 / 203.96 ±2.74 / 208.69 ms │                          200.37 / 204.51 ±3.79 / 209.59 ms │     no change │
│ QQuery 79 │        68.45 / 69.96 ±1.81 / 73.29 ms │                             68.61 / 69.10 ±0.61 / 70.26 ms │     no change │
│ QQuery 80 │      99.86 / 101.45 ±1.21 / 103.40 ms │                          101.45 / 102.13 ±1.21 / 104.54 ms │     no change │
│ QQuery 81 │        26.44 / 26.69 ±0.19 / 26.96 ms │                             26.49 / 26.85 ±0.38 / 27.55 ms │     no change │
│ QQuery 82 │        16.88 / 17.07 ±0.15 / 17.31 ms │                             16.68 / 16.86 ±0.11 / 17.00 ms │     no change │
│ QQuery 83 │        40.54 / 40.82 ±0.29 / 41.27 ms │                             40.82 / 41.15 ±0.39 / 41.84 ms │     no change │
│ QQuery 84 │        30.97 / 33.77 ±3.68 / 40.79 ms │                             30.57 / 30.92 ±0.22 / 31.25 ms │ +1.09x faster │
│ QQuery 85 │     108.02 / 108.37 ±0.33 / 108.92 ms │                          110.06 / 113.37 ±4.74 / 122.63 ms │     no change │
│ QQuery 86 │        25.84 / 25.93 ±0.10 / 26.10 ms │                             26.03 / 26.42 ±0.44 / 27.27 ms │     no change │
│ QQuery 87 │        64.46 / 66.14 ±1.67 / 69.04 ms │                             64.75 / 66.24 ±0.98 / 67.48 ms │     no change │
│ QQuery 88 │        63.44 / 63.69 ±0.25 / 64.13 ms │                             63.97 / 67.07 ±5.53 / 78.11 ms │  1.05x slower │
│ QQuery 89 │        35.92 / 36.19 ±0.18 / 36.40 ms │                             36.33 / 36.92 ±0.89 / 38.68 ms │     no change │
│ QQuery 90 │        17.63 / 17.84 ±0.17 / 18.01 ms │                             17.47 / 17.66 ±0.23 / 18.11 ms │     no change │
│ QQuery 91 │        45.80 / 46.08 ±0.31 / 46.60 ms │                             46.36 / 47.01 ±0.37 / 47.35 ms │     no change │
│ QQuery 92 │        30.97 / 32.55 ±1.86 / 35.34 ms │                             30.52 / 31.15 ±0.42 / 31.79 ms │     no change │
│ QQuery 93 │        50.22 / 51.16 ±0.80 / 52.11 ms │                             51.45 / 52.63 ±1.39 / 55.13 ms │     no change │
│ QQuery 94 │        39.07 / 39.30 ±0.18 / 39.56 ms │                             39.33 / 40.06 ±0.55 / 40.85 ms │     no change │
│ QQuery 95 │        80.97 / 81.47 ±0.36 / 82.02 ms │                             81.58 / 82.37 ±0.54 / 83.13 ms │     no change │
│ QQuery 96 │        24.88 / 25.78 ±1.22 / 28.18 ms │                             24.71 / 24.85 ±0.15 / 25.08 ms │     no change │
│ QQuery 97 │        47.39 / 48.23 ±1.02 / 50.16 ms │                             47.92 / 49.70 ±2.29 / 54.14 ms │     no change │
│ QQuery 98 │        43.43 / 43.57 ±0.17 / 43.90 ms │                             43.74 / 44.38 ±0.40 / 44.85 ms │     no change │
│ QQuery 99 │        70.52 / 71.18 ±0.41 / 71.64 ms │                             70.80 / 71.43 ±0.64 / 72.67 ms │     no change │
└───────────┴───────────────────────────────────────┴────────────────────────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                                         ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                                         │ 10001.90ms │
│ Total Time (xudong963_fix-interleaved-hash-join-sort-pushdown-upstream)   │ 10177.03ms │
│ Average Time (HEAD)                                                       │   101.03ms │
│ Average Time (xudong963_fix-interleaved-hash-join-sort-pushdown-upstream) │   102.80ms │
│ Queries Faster                                                            │          2 │
│ Queries Slower                                                            │          2 │
│ Queries with No Change                                                    │         95 │
│ Queries with Failure                                                      │          0 │
└───────────────────────────────────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 55.0s
Peak memory 2.2 GiB
Avg memory 1.5 GiB
CPU user 223.4s
CPU sys 6.2s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 55.0s
Peak memory 2.1 GiB
Avg memory 1.4 GiB
CPU user 232.9s
CPU sys 6.0s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 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 xudong963_fix-interleaved-hash-join-sort-pushdown-upstream
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ xudong963_fix-interleaved-hash-join-sort-pushdown-upstream ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.23 / 4.10 ±5.56 / 15.22 ms │                               1.26 / 4.12 ±5.57 / 15.26 ms │    no change │
│ QQuery 1  │        13.12 / 13.43 ±0.20 / 13.70 ms │                             13.38 / 13.63 ±0.17 / 13.88 ms │    no change │
│ QQuery 2  │        36.46 / 36.74 ±0.31 / 37.25 ms │                             36.58 / 36.77 ±0.23 / 37.21 ms │    no change │
│ QQuery 3  │        31.14 / 31.65 ±0.57 / 32.76 ms │                             30.84 / 31.21 ±0.28 / 31.50 ms │    no change │
│ QQuery 4  │     232.26 / 233.57 ±0.91 / 234.62 ms │                          228.06 / 233.06 ±4.77 / 242.13 ms │    no change │
│ QQuery 5  │     277.16 / 278.33 ±0.69 / 278.98 ms │                          274.16 / 277.94 ±3.09 / 282.63 ms │    no change │
│ QQuery 6  │           1.30 / 1.46 ±0.23 / 1.91 ms │                                1.28 / 1.46 ±0.23 / 1.89 ms │    no change │
│ QQuery 7  │        14.71 / 14.87 ±0.17 / 15.20 ms │                             14.31 / 14.52 ±0.13 / 14.67 ms │    no change │
│ QQuery 8  │     334.87 / 340.52 ±4.06 / 346.38 ms │                          328.71 / 334.56 ±4.75 / 342.55 ms │    no change │
│ QQuery 9  │     472.00 / 475.47 ±4.51 / 484.25 ms │                          454.33 / 462.72 ±7.64 / 473.57 ms │    no change │
│ QQuery 10 │        71.09 / 72.19 ±1.63 / 75.43 ms │                             71.55 / 72.61 ±0.58 / 73.21 ms │    no change │
│ QQuery 11 │        83.29 / 86.00 ±4.27 / 94.48 ms │                             83.56 / 84.27 ±0.50 / 85.02 ms │    no change │
│ QQuery 12 │     274.25 / 279.65 ±4.42 / 285.09 ms │                          270.83 / 277.55 ±6.28 / 286.96 ms │    no change │
│ QQuery 13 │     376.28 / 381.36 ±3.89 / 387.20 ms │                         368.20 / 391.15 ±18.13 / 415.99 ms │    no change │
│ QQuery 14 │     285.56 / 293.13 ±4.62 / 299.89 ms │                          289.60 / 298.56 ±5.70 / 303.41 ms │    no change │
│ QQuery 15 │     274.31 / 280.35 ±4.16 / 286.87 ms │                          278.31 / 285.36 ±6.16 / 293.62 ms │    no change │
│ QQuery 16 │     621.80 / 628.91 ±5.14 / 635.02 ms │                          631.26 / 641.10 ±9.20 / 654.30 ms │    no change │
│ QQuery 17 │    629.32 / 642.37 ±12.40 / 660.37 ms │                         645.02 / 661.69 ±12.69 / 674.00 ms │    no change │
│ QQuery 18 │  1308.15 / 1316.03 ±7.89 / 1328.80 ms │                      1310.49 / 1352.02 ±29.92 / 1399.82 ms │    no change │
│ QQuery 19 │        28.40 / 28.67 ±0.15 / 28.82 ms │                            29.28 / 45.08 ±25.47 / 95.31 ms │ 1.57x slower │
│ QQuery 20 │    515.86 / 546.92 ±37.25 / 613.21 ms │                         525.51 / 536.96 ±11.85 / 556.09 ms │    no change │
│ QQuery 21 │     513.64 / 521.32 ±4.55 / 525.41 ms │                          530.16 / 533.57 ±2.92 / 538.18 ms │    no change │
│ QQuery 22 │    988.15 / 996.21 ±7.63 / 1008.83 ms │                       1026.32 / 1033.77 ±6.57 / 1043.58 ms │    no change │
│ QQuery 23 │ 3056.02 / 3113.00 ±36.28 / 3168.34 ms │                      3258.39 / 3304.83 ±35.00 / 3362.86 ms │ 1.06x slower │
│ QQuery 24 │        42.09 / 44.09 ±1.66 / 46.75 ms │                             42.60 / 50.01 ±9.61 / 67.36 ms │ 1.13x slower │
│ QQuery 25 │     110.99 / 112.41 ±1.35 / 114.87 ms │                          118.42 / 119.24 ±0.66 / 120.45 ms │ 1.06x slower │
│ QQuery 26 │        41.76 / 43.26 ±1.77 / 46.12 ms │                             44.09 / 46.26 ±2.68 / 51.33 ms │ 1.07x slower │
│ QQuery 27 │    519.39 / 530.80 ±15.03 / 560.31 ms │                          533.00 / 544.89 ±8.14 / 552.87 ms │    no change │
│ QQuery 28 │ 2875.94 / 2901.96 ±17.89 / 2924.12 ms │                      3026.75 / 3064.91 ±23.69 / 3094.94 ms │ 1.06x slower │
│ QQuery 29 │       41.82 / 52.26 ±15.15 / 81.07 ms │                            43.28 / 56.01 ±10.99 / 70.63 ms │ 1.07x slower │
│ QQuery 30 │     307.29 / 315.83 ±5.82 / 323.11 ms │                          340.70 / 344.61 ±2.85 / 348.40 ms │ 1.09x slower │
│ QQuery 31 │     295.58 / 303.06 ±5.13 / 311.13 ms │                          310.62 / 316.99 ±4.99 / 324.52 ms │    no change │
│ QQuery 32 │    949.94 / 970.25 ±15.57 / 989.19 ms │                      1047.61 / 1091.90 ±27.50 / 1124.74 ms │ 1.13x slower │
│ QQuery 33 │ 1466.45 / 1492.74 ±15.19 / 1511.00 ms │                      1635.56 / 1673.83 ±25.85 / 1713.66 ms │ 1.12x slower │
│ QQuery 34 │ 1504.03 / 1556.90 ±48.10 / 1634.38 ms │                      1607.29 / 1658.69 ±33.19 / 1708.58 ms │ 1.07x slower │
│ QQuery 35 │    280.14 / 309.96 ±37.51 / 383.21 ms │                         313.70 / 341.75 ±29.50 / 393.09 ms │ 1.10x slower │
│ QQuery 36 │        70.65 / 73.97 ±3.22 / 78.65 ms │                             71.38 / 76.19 ±4.28 / 81.43 ms │    no change │
│ QQuery 37 │        36.19 / 38.38 ±2.40 / 43.05 ms │                             37.56 / 43.69 ±3.38 / 46.97 ms │ 1.14x slower │
│ QQuery 38 │        40.50 / 41.55 ±0.92 / 42.92 ms │                             41.17 / 47.22 ±6.72 / 60.16 ms │ 1.14x slower │
│ QQuery 39 │     155.09 / 158.29 ±3.02 / 163.99 ms │                         149.89 / 166.82 ±11.67 / 182.48 ms │ 1.05x slower │
│ QQuery 40 │        14.74 / 15.35 ±0.53 / 16.23 ms │                             15.69 / 19.07 ±4.32 / 27.59 ms │ 1.24x slower │
│ QQuery 41 │        14.06 / 14.29 ±0.12 / 14.40 ms │                             14.92 / 16.02 ±1.55 / 19.03 ms │ 1.12x slower │
│ QQuery 42 │        13.67 / 13.81 ±0.11 / 13.95 ms │                             14.41 / 14.54 ±0.12 / 14.75 ms │ 1.05x slower │
└───────────┴───────────────────────────────────────┴────────────────────────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                                         ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                                         │ 19605.42ms │
│ Total Time (xudong963_fix-interleaved-hash-join-sort-pushdown-upstream)   │ 20621.15ms │
│ Average Time (HEAD)                                                       │   455.94ms │
│ Average Time (xudong963_fix-interleaved-hash-join-sort-pushdown-upstream) │   479.56ms │
│ Queries Faster                                                            │          0 │
│ Queries Slower                                                            │         18 │
│ Queries with No Change                                                    │         25 │
│ Queries with Failure                                                      │          0 │
└───────────────────────────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 100.0s
Peak memory 12.0 GiB
Avg memory 4.3 GiB
CPU user 1008.1s
CPU sys 69.5s
Peak spill 0 B

clickbench_partitioned — branch

Metric Value
Wall time 105.0s
Peak memory 11.5 GiB
Avg memory 4.6 GiB
CPU user 1054.5s
CPU sys 79.3s
Peak spill 0 B

File an issue against this benchmark runner

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

core Core DataFusion crate optimizer Optimizer rules

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Incorrect sort pushdown through interleaved HashJoin projection

3 participants