Add eager group join strategy by nathanb9 · Pull Request #22058 · apache/datafusion

nathanb9 · 2026-05-07T05:28:23Z

This adds an eager right-side aggregation strategy for GroupJoin. The optimizer selects this strategy when row-count estimates suggest it is cheaper to aggregate the right side up front, then probe those precomputed aggregate results from the left side.

The eager strategy triggers in two cases:

Left join: right_rows > 1.5 * left_rows
Inner join: right_rows > 2 * left_rows
Execution aggregates the right side by join key first, evaluates aggregate accumulators once, then uses indexed lookups while scanning left-side batches. This avoids repeated accumulator updates during the probe phase and keeps left-side output pipelined.

github-actions · 2026-05-07T05:41:13Z

Thank you for opening this pull request!

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

Details

     Cloning apache/main
    Building datafusion v53.1.0 (current)
       Built [  83.726s] (current)
     Parsing datafusion v53.1.0 (current)
      Parsed [   0.034s] (current)
    Building datafusion v53.1.0 (baseline)
       Built [  82.101s] (baseline)
     Parsing datafusion v53.1.0 (baseline)
      Parsed [   0.034s] (baseline)
    Checking datafusion v53.1.0 -> v53.1.0 (no change; assume patch)
     Checked [   0.621s] 222 checks: 222 pass, 30 skip
     Summary no semver update required
    Finished [ 168.005s] datafusion
    Building datafusion-physical-optimizer v53.1.0 (current)
       Built [  37.433s] (current)
     Parsing datafusion-physical-optimizer v53.1.0 (current)
      Parsed [   0.022s] (current)
    Building datafusion-physical-optimizer v53.1.0 (baseline)
       Built [  37.230s] (baseline)
     Parsing datafusion-physical-optimizer v53.1.0 (baseline)
      Parsed [   0.023s] (baseline)
    Checking datafusion-physical-optimizer v53.1.0 -> v53.1.0 (no change; assume patch)
     Checked [   0.124s] 222 checks: 222 pass, 30 skip
     Summary no semver update required
    Finished [  76.198s] datafusion-physical-optimizer
    Building datafusion-physical-plan v53.1.0 (current)
       Built [  32.291s] (current)
     Parsing datafusion-physical-plan v53.1.0 (current)
      Parsed [   0.124s] (current)
    Building datafusion-physical-plan v53.1.0 (baseline)
       Built [  32.831s] (baseline)
     Parsing datafusion-physical-plan v53.1.0 (baseline)
      Parsed [   0.128s] (baseline)
    Checking datafusion-physical-plan v53.1.0 -> v53.1.0 (no change; assume patch)
     Checked [   0.647s] 222 checks: 222 pass, 30 skip
     Summary no semver update required
    Finished [  67.374s] datafusion-physical-plan
    Building datafusion-proto v53.1.0 (current)
       Built [  54.695s] (current)
     Parsing datafusion-proto v53.1.0 (current)
      Parsed [   0.133s] (current)
    Building datafusion-proto v53.1.0 (baseline)
       Built [  54.372s] (baseline)
     Parsing datafusion-proto v53.1.0 (baseline)
      Parsed [   0.137s] (baseline)
    Checking datafusion-proto v53.1.0 -> v53.1.0 (no change; assume patch)
     Checked [   1.897s] 222 checks: 221 pass, 1 fail, 0 warn, 30 skip

--- failure enum_variant_added: enum variant added on exhaustive enum ---

Description:
A publicly-visible enum without #[non_exhaustive] has a new variant.
        ref: https://doc.rust-lang.org/cargo/reference/semver.html#enum-variant-new
       impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.47.0/src/lints/enum_variant_added.ron

Failed in:
  variant PhysicalPlanType:GroupJoin in /home/runner/work/datafusion/datafusion/datafusion/proto/src/generated/prost.rs:1209
  variant PhysicalPlanType:EagerRightGroupJoin in /home/runner/work/datafusion/datafusion/datafusion/proto/src/generated/prost.rs:1211
  variant PhysicalPlanType:GroupJoin in /home/runner/work/datafusion/datafusion/datafusion/proto/src/generated/prost.rs:1209
  variant PhysicalPlanType:EagerRightGroupJoin in /home/runner/work/datafusion/datafusion/datafusion/proto/src/generated/prost.rs:1211

     Summary semver requires new major version: 1 major and 0 minor checks failed
    Finished [ 113.008s] datafusion-proto
    Building datafusion-sqllogictest v53.1.0 (current)
       Built [ 135.692s] (current)
     Parsing datafusion-sqllogictest v53.1.0 (current)
      Parsed [   0.021s] (current)
    Building datafusion-sqllogictest v53.1.0 (baseline)
       Built [ 134.833s] (baseline)
     Parsing datafusion-sqllogictest v53.1.0 (baseline)
      Parsed [   0.022s] (baseline)
    Checking datafusion-sqllogictest v53.1.0 -> v53.1.0 (no change; assume patch)
     Checked [   0.095s] 222 checks: 222 pass, 30 skip
     Summary no semver update required
    Finished [ 273.504s] datafusion-sqllogictest

nathanb9 · 2026-05-07T13:59:22Z

@Dandandan Could you please help me run the benchmarks here?

Dandandan · 2026-05-07T14:05:45Z

run benchmark tpch tpch10 tpcds

adriangbot · 2026-05-07T14:09:09Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4397823269-2052-4x47x 6.12.68+ #1 SMP Wed Apr 1 02:23:28 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 groupjoin-eager-strategy (30075d5) to 2f2fe8f (merge-base) diff using: tpcds
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-05-07T14:09:11Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4397823269-2050-lr6ld 6.12.68+ #1 SMP Wed Apr 1 02:23:28 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 groupjoin-eager-strategy (30075d5) to 2f2fe8f (merge-base) diff using: tpch
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-05-07T14:09:21Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4397823269-2051-8vmnr 6.12.68+ #1 SMP Wed Apr 1 02:23:28 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 groupjoin-eager-strategy (30075d5) to 2f2fe8f (merge-base) diff using: tpch10
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-05-07T14:22:18Z

🤖 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 groupjoin-eager-strategy
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃       groupjoin-eager-strategy ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 40.64 / 41.57 ±0.92 / 43.23 ms │ 40.16 / 40.81 ±1.20 / 43.21 ms │     no change │
│ QQuery 2  │ 21.50 / 22.45 ±0.82 / 23.47 ms │ 21.29 / 21.62 ±0.37 / 22.21 ms │     no change │
│ QQuery 3  │ 36.04 / 37.88 ±1.44 / 39.68 ms │ 36.01 / 38.38 ±2.41 / 42.61 ms │     no change │
│ QQuery 4  │ 18.39 / 18.53 ±0.14 / 18.75 ms │ 18.39 / 18.98 ±0.66 / 20.14 ms │     no change │
│ QQuery 5  │ 43.72 / 46.22 ±1.39 / 47.73 ms │ 43.98 / 45.76 ±1.39 / 47.13 ms │     no change │
│ QQuery 6  │ 17.27 / 18.26 ±1.15 / 20.52 ms │ 17.50 / 17.58 ±0.07 / 17.69 ms │     no change │
│ QQuery 7  │ 51.86 / 53.42 ±1.95 / 56.89 ms │ 51.33 / 52.11 ±0.86 / 53.51 ms │     no change │
│ QQuery 8  │ 47.28 / 47.60 ±0.17 / 47.74 ms │ 46.91 / 47.41 ±0.53 / 48.34 ms │     no change │
│ QQuery 9  │ 51.49 / 52.61 ±0.58 / 53.06 ms │ 52.21 / 52.97 ±0.66 / 53.87 ms │     no change │
│ QQuery 10 │ 66.23 / 66.75 ±0.85 / 68.44 ms │ 65.57 / 66.51 ±0.79 / 67.46 ms │     no change │
│ QQuery 11 │ 13.76 / 14.12 ±0.22 / 14.45 ms │ 13.92 / 14.26 ±0.38 / 14.99 ms │     no change │
│ QQuery 12 │ 26.32 / 26.84 ±0.34 / 27.37 ms │ 26.23 / 26.52 ±0.21 / 26.76 ms │     no change │
│ QQuery 13 │ 35.79 / 36.90 ±0.65 / 37.71 ms │ 33.67 / 34.27 ±0.56 / 35.31 ms │ +1.08x faster │
│ QQuery 14 │ 27.04 / 27.19 ±0.10 / 27.32 ms │ 26.73 / 26.90 ±0.23 / 27.36 ms │     no change │
│ QQuery 15 │ 33.34 / 33.94 ±0.78 / 35.48 ms │ 33.17 / 33.71 ±0.65 / 34.95 ms │     no change │
│ QQuery 16 │ 15.54 / 15.63 ±0.10 / 15.81 ms │ 15.31 / 15.50 ±0.12 / 15.62 ms │     no change │
│ QQuery 17 │ 78.06 / 79.28 ±1.34 / 81.58 ms │ 76.86 / 78.22 ±1.05 / 79.60 ms │     no change │
│ QQuery 18 │ 70.44 / 72.61 ±2.38 / 77.18 ms │ 69.40 / 70.56 ±0.69 / 71.44 ms │     no change │
│ QQuery 19 │ 37.61 / 38.09 ±0.28 / 38.39 ms │ 37.43 / 37.71 ±0.17 / 37.96 ms │     no change │
│ QQuery 20 │ 39.49 / 39.69 ±0.19 / 39.98 ms │ 39.31 / 39.56 ±0.35 / 40.23 ms │     no change │
│ QQuery 21 │ 61.67 / 64.08 ±2.09 / 67.60 ms │ 59.39 / 61.58 ±1.61 / 64.06 ms │     no change │
│ QQuery 22 │ 24.19 / 24.52 ±0.29 / 24.91 ms │ 24.44 / 24.57 ±0.14 / 24.83 ms │     no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                       ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                       │ 878.19ms │
│ Total Time (groupjoin-eager-strategy)   │ 865.50ms │
│ Average Time (HEAD)                     │  39.92ms │
│ Average Time (groupjoin-eager-strategy) │  39.34ms │
│ Queries Faster                          │        1 │
│ Queries Slower                          │        0 │
│ Queries with No Change                  │       21 │
│ Queries with Failure                    │        0 │
└─────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric	Value
Wall time	5.0s
Peak memory	5.7 GiB
Avg memory	5.0 GiB
CPU user	32.5s
CPU sys	2.4s
Peak spill	0 B

tpch — branch

Metric	Value
Wall time	5.0s
Peak memory	5.4 GiB
Avg memory	5.0 GiB
CPU user	32.2s
CPU sys	2.3s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-05-07T14:24:10Z

🤖 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 groupjoin-eager-strategy
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃              groupjoin-eager-strategy ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           6.23 / 6.73 ±0.89 / 8.50 ms │           6.43 / 7.00 ±0.84 / 8.66 ms │     no change │
│ QQuery 2  │        81.72 / 81.89 ±0.10 / 82.01 ms │        84.18 / 84.35 ±0.16 / 84.64 ms │     no change │
│ QQuery 3  │        29.84 / 30.27 ±0.34 / 30.83 ms │        31.05 / 31.31 ±0.27 / 31.70 ms │     no change │
│ QQuery 4  │     527.78 / 535.11 ±4.89 / 542.95 ms │     546.20 / 550.20 ±2.97 / 555.26 ms │     no change │
│ QQuery 5  │        52.76 / 53.85 ±0.59 / 54.53 ms │        54.83 / 55.16 ±0.34 / 55.72 ms │     no change │
│ QQuery 6  │        36.72 / 36.80 ±0.10 / 36.98 ms │        37.58 / 38.09 ±0.48 / 38.99 ms │     no change │
│ QQuery 7  │     112.11 / 114.77 ±2.95 / 120.40 ms │     115.28 / 121.16 ±3.82 / 127.34 ms │  1.06x slower │
│ QQuery 8  │        39.60 / 39.81 ±0.14 / 40.02 ms │        40.64 / 40.78 ±0.13 / 41.00 ms │     no change │
│ QQuery 9  │        55.97 / 56.88 ±0.84 / 58.21 ms │        57.71 / 58.68 ±0.91 / 60.37 ms │     no change │
│ QQuery 10 │        84.07 / 84.50 ±0.37 / 85.19 ms │        85.31 / 87.15 ±1.46 / 89.56 ms │     no change │
│ QQuery 11 │     320.52 / 324.72 ±3.96 / 331.31 ms │     331.65 / 335.21 ±3.74 / 341.60 ms │     no change │
│ QQuery 12 │        28.90 / 29.49 ±0.41 / 30.08 ms │        29.89 / 30.25 ±0.34 / 30.81 ms │     no change │
│ QQuery 13 │     132.07 / 133.92 ±1.44 / 136.50 ms │     132.64 / 133.88 ±0.89 / 135.18 ms │     no change │
│ QQuery 14 │     517.37 / 520.68 ±1.71 / 522.27 ms │     527.03 / 528.64 ±1.21 / 530.32 ms │     no change │
│ QQuery 15 │        60.54 / 62.20 ±1.23 / 63.83 ms │        62.30 / 63.56 ±1.41 / 65.60 ms │     no change │
│ QQuery 16 │           6.94 / 7.11 ±0.18 / 7.43 ms │           6.79 / 6.98 ±0.24 / 7.45 ms │     no change │
│ QQuery 17 │        83.89 / 85.80 ±1.75 / 88.01 ms │        84.56 / 87.13 ±2.12 / 90.17 ms │     no change │
│ QQuery 18 │     156.72 / 158.15 ±1.24 / 160.16 ms │     157.80 / 158.76 ±0.95 / 160.45 ms │     no change │
│ QQuery 19 │        42.31 / 42.64 ±0.23 / 42.94 ms │        42.99 / 43.33 ±0.23 / 43.63 ms │     no change │
│ QQuery 20 │        36.12 / 36.69 ±0.57 / 37.72 ms │        36.64 / 37.08 ±0.47 / 37.93 ms │     no change │
│ QQuery 21 │        18.29 / 18.60 ±0.19 / 18.79 ms │        18.76 / 19.03 ±0.15 / 19.19 ms │     no change │
│ QQuery 22 │        61.80 / 62.34 ±0.36 / 62.89 ms │        62.21 / 63.46 ±1.79 / 67.01 ms │     no change │
│ QQuery 23 │     487.14 / 489.74 ±2.82 / 495.17 ms │     486.67 / 493.56 ±7.22 / 507.28 ms │     no change │
│ QQuery 24 │     246.53 / 248.46 ±1.33 / 250.45 ms │     248.47 / 250.92 ±2.45 / 255.37 ms │     no change │
│ QQuery 25 │     117.38 / 118.14 ±0.51 / 118.84 ms │     119.24 / 120.79 ±1.58 / 123.82 ms │     no change │
│ QQuery 26 │        74.25 / 74.78 ±0.72 / 76.11 ms │        74.68 / 75.66 ±1.43 / 78.48 ms │     no change │
│ QQuery 27 │          7.13 / 7.96 ±1.40 / 10.74 ms │           7.03 / 7.16 ±0.15 / 7.41 ms │ +1.11x faster │
│ QQuery 28 │        65.21 / 65.56 ±0.26 / 66.00 ms │        66.61 / 67.21 ±0.40 / 67.75 ms │     no change │
│ QQuery 29 │     101.54 / 102.63 ±1.08 / 104.50 ms │     104.52 / 107.65 ±3.10 / 112.98 ms │     no change │
│ QQuery 30 │                                  FAIL │                                  FAIL │  incomparable │
│ QQuery 31 │     114.86 / 116.24 ±1.57 / 119.25 ms │     117.06 / 118.03 ±0.61 / 118.69 ms │     no change │
│ QQuery 32 │        20.68 / 20.96 ±0.16 / 21.18 ms │        21.60 / 21.74 ±0.20 / 22.14 ms │     no change │
│ QQuery 33 │        39.92 / 40.40 ±0.44 / 41.19 ms │        40.98 / 41.42 ±0.31 / 41.92 ms │     no change │
│ QQuery 34 │        10.09 / 10.91 ±1.05 / 12.94 ms │        10.53 / 11.26 ±1.01 / 13.25 ms │     no change │
│ QQuery 35 │        82.57 / 83.22 ±0.43 / 83.81 ms │        83.23 / 83.99 ±0.82 / 85.45 ms │     no change │
│ QQuery 36 │           6.54 / 6.64 ±0.16 / 6.96 ms │           6.53 / 6.67 ±0.17 / 6.96 ms │     no change │
│ QQuery 37 │           7.18 / 7.31 ±0.07 / 7.37 ms │           7.27 / 7.37 ±0.12 / 7.59 ms │     no change │
│ QQuery 38 │        67.93 / 68.73 ±0.83 / 70.20 ms │        69.27 / 69.94 ±0.38 / 70.44 ms │     no change │
│ QQuery 39 │     100.32 / 101.32 ±0.71 / 102.47 ms │     101.98 / 103.71 ±1.89 / 107.13 ms │     no change │
│ QQuery 40 │        23.50 / 23.65 ±0.11 / 23.81 ms │        23.78 / 24.02 ±0.17 / 24.28 ms │     no change │
│ QQuery 41 │        14.00 / 14.10 ±0.15 / 14.40 ms │        14.18 / 14.70 ±0.73 / 16.15 ms │     no change │
│ QQuery 42 │        24.96 / 25.17 ±0.15 / 25.30 ms │        25.51 / 25.96 ±0.37 / 26.48 ms │     no change │
│ QQuery 43 │           5.19 / 5.27 ±0.08 / 5.42 ms │           5.32 / 5.46 ±0.16 / 5.77 ms │     no change │
│ QQuery 44 │        10.86 / 10.89 ±0.04 / 10.96 ms │        11.00 / 11.18 ±0.11 / 11.32 ms │     no change │
│ QQuery 45 │        40.95 / 41.56 ±0.41 / 41.97 ms │        41.70 / 42.74 ±0.70 / 43.60 ms │     no change │
│ QQuery 46 │        13.79 / 14.27 ±0.82 / 15.90 ms │        14.28 / 14.58 ±0.39 / 15.33 ms │     no change │
│ QQuery 47 │     234.64 / 235.79 ±1.10 / 237.77 ms │     237.31 / 239.09 ±1.23 / 240.52 ms │     no change │
│ QQuery 48 │     105.67 / 107.51 ±1.50 / 110.11 ms │     106.92 / 107.63 ±0.80 / 109.19 ms │     no change │
│ QQuery 49 │        82.63 / 84.28 ±0.99 / 85.69 ms │        84.53 / 85.38 ±0.53 / 86.08 ms │     no change │
│ QQuery 50 │        61.75 / 63.71 ±2.98 / 69.63 ms │        62.85 / 64.34 ±1.71 / 66.52 ms │     no change │
│ QQuery 51 │        94.37 / 95.98 ±1.51 / 98.39 ms │        95.49 / 97.18 ±1.12 / 98.71 ms │     no change │
│ QQuery 52 │        25.35 / 26.13 ±0.91 / 27.92 ms │        25.45 / 25.99 ±0.38 / 26.43 ms │     no change │
│ QQuery 53 │        30.98 / 31.27 ±0.19 / 31.54 ms │        31.70 / 31.86 ±0.20 / 32.24 ms │     no change │
│ QQuery 54 │        55.46 / 56.18 ±0.50 / 56.88 ms │        55.77 / 56.86 ±1.60 / 59.98 ms │     no change │
│ QQuery 55 │        25.22 / 25.39 ±0.19 / 25.76 ms │        24.98 / 25.42 ±0.25 / 25.69 ms │     no change │
│ QQuery 56 │        40.77 / 41.28 ±0.42 / 41.83 ms │        41.23 / 41.59 ±0.33 / 42.11 ms │     no change │
│ QQuery 57 │     179.23 / 181.45 ±1.70 / 184.25 ms │     183.42 / 185.53 ±1.30 / 187.45 ms │     no change │
│ QQuery 58 │     117.68 / 118.60 ±0.77 / 119.49 ms │     121.72 / 122.54 ±0.66 / 123.30 ms │     no change │
│ QQuery 59 │     118.35 / 118.89 ±0.42 / 119.48 ms │     119.45 / 120.02 ±0.50 / 120.68 ms │     no change │
│ QQuery 60 │        40.72 / 41.58 ±0.88 / 43.25 ms │        41.22 / 42.12 ±0.65 / 42.98 ms │     no change │
│ QQuery 61 │        13.46 / 13.60 ±0.17 / 13.93 ms │        13.95 / 14.07 ±0.08 / 14.17 ms │     no change │
│ QQuery 62 │        47.72 / 48.51 ±0.86 / 50.17 ms │        48.68 / 49.85 ±1.81 / 53.46 ms │     no change │
│ QQuery 63 │        30.84 / 31.27 ±0.24 / 31.52 ms │        32.08 / 32.57 ±0.28 / 32.82 ms │     no change │
│ QQuery 64 │     478.46 / 483.45 ±5.03 / 492.52 ms │     484.56 / 487.14 ±1.52 / 488.95 ms │     no change │
│ QQuery 65 │     145.10 / 149.21 ±3.57 / 155.20 ms │     146.41 / 151.50 ±3.37 / 155.30 ms │     no change │
│ QQuery 66 │        84.53 / 86.52 ±2.92 / 92.29 ms │        84.82 / 86.61 ±2.26 / 90.98 ms │     no change │
│ QQuery 67 │     245.48 / 250.66 ±4.70 / 257.95 ms │     255.42 / 259.17 ±2.11 / 260.78 ms │     no change │
│ QQuery 68 │        13.81 / 14.03 ±0.21 / 14.44 ms │        14.15 / 14.38 ±0.24 / 14.83 ms │     no change │
│ QQuery 69 │        80.15 / 80.28 ±0.18 / 80.62 ms │        80.69 / 82.66 ±3.32 / 89.30 ms │     no change │
│ QQuery 70 │     106.74 / 111.20 ±5.59 / 120.82 ms │     107.91 / 112.43 ±4.36 / 117.92 ms │     no change │
│ QQuery 71 │        36.63 / 37.24 ±0.34 / 37.55 ms │        37.63 / 37.95 ±0.26 / 38.36 ms │     no change │
│ QQuery 72 │ 2245.79 / 2315.60 ±55.29 / 2391.78 ms │ 2209.14 / 2269.82 ±46.57 / 2342.41 ms │     no change │
│ QQuery 73 │        10.35 / 10.54 ±0.11 / 10.68 ms │         9.98 / 10.28 ±0.25 / 10.73 ms │     no change │
│ QQuery 74 │     191.35 / 199.68 ±4.28 / 203.26 ms │     183.68 / 186.68 ±2.58 / 190.95 ms │ +1.07x faster │
│ QQuery 75 │     154.51 / 156.01 ±0.86 / 157.18 ms │     154.56 / 156.66 ±2.77 / 162.05 ms │     no change │
│ QQuery 76 │        37.07 / 37.35 ±0.28 / 37.73 ms │        37.65 / 38.15 ±0.34 / 38.60 ms │     no change │
│ QQuery 77 │        64.10 / 64.63 ±0.67 / 65.94 ms │        64.47 / 65.00 ±0.52 / 65.97 ms │     no change │
│ QQuery 78 │     193.90 / 200.21 ±3.92 / 206.00 ms │     196.20 / 201.09 ±6.18 / 213.21 ms │     no change │
│ QQuery 79 │        69.96 / 70.36 ±0.23 / 70.68 ms │        69.20 / 69.76 ±0.30 / 70.12 ms │     no change │
│ QQuery 80 │     108.95 / 110.22 ±0.75 / 111.07 ms │     104.73 / 109.44 ±3.69 / 115.96 ms │     no change │
│ QQuery 81 │        25.62 / 25.91 ±0.18 / 26.20 ms │        25.68 / 25.77 ±0.10 / 25.94 ms │     no change │
│ QQuery 82 │        18.25 / 18.53 ±0.18 / 18.75 ms │        17.99 / 18.84 ±1.42 / 21.66 ms │     no change │
│ QQuery 83 │        38.81 / 39.36 ±0.38 / 39.95 ms │        38.76 / 39.08 ±0.36 / 39.53 ms │     no change │
│ QQuery 84 │        45.08 / 46.05 ±0.76 / 47.39 ms │        44.08 / 44.44 ±0.31 / 44.93 ms │     no change │
│ QQuery 85 │     142.78 / 144.13 ±0.82 / 145.14 ms │     141.32 / 142.64 ±1.35 / 144.98 ms │     no change │
│ QQuery 86 │        26.96 / 27.12 ±0.11 / 27.29 ms │        25.63 / 26.13 ±0.38 / 26.66 ms │     no change │
│ QQuery 87 │        72.87 / 74.54 ±1.47 / 76.77 ms │        69.65 / 70.58 ±1.33 / 73.23 ms │ +1.06x faster │
│ QQuery 88 │        66.51 / 67.11 ±0.45 / 67.87 ms │        66.04 / 66.12 ±0.07 / 66.20 ms │     no change │
│ QQuery 89 │        37.81 / 38.49 ±0.41 / 39.06 ms │        37.34 / 38.09 ±0.67 / 39.24 ms │     no change │
│ QQuery 90 │        18.13 / 18.32 ±0.17 / 18.61 ms │        18.09 / 18.27 ±0.19 / 18.56 ms │     no change │
│ QQuery 91 │        55.21 / 55.98 ±0.82 / 57.49 ms │        53.72 / 54.33 ±0.60 / 55.36 ms │     no change │
│ QQuery 92 │        31.72 / 32.58 ±0.54 / 33.32 ms │        30.95 / 31.67 ±0.56 / 32.17 ms │     no change │
│ QQuery 93 │        52.83 / 54.45 ±1.48 / 57.18 ms │        53.02 / 53.71 ±1.01 / 55.66 ms │     no change │
│ QQuery 94 │        41.39 / 41.63 ±0.31 / 42.19 ms │        41.17 / 41.44 ±0.15 / 41.59 ms │     no change │
│ QQuery 95 │        90.10 / 90.99 ±0.95 / 92.74 ms │        89.76 / 90.68 ±0.66 / 91.81 ms │     no change │
│ QQuery 96 │        25.11 / 25.41 ±0.20 / 25.74 ms │        24.85 / 25.12 ±0.23 / 25.49 ms │     no change │
│ QQuery 97 │        48.47 / 48.82 ±0.35 / 49.44 ms │        48.02 / 48.38 ±0.34 / 48.88 ms │     no change │
│ QQuery 98 │        43.52 / 44.09 ±0.50 / 44.79 ms │        43.83 / 44.37 ±0.47 / 45.17 ms │     no change │
│ QQuery 99 │        72.23 / 72.75 ±0.49 / 73.58 ms │        71.82 / 72.47 ±0.78 / 73.99 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                       ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                       │ 10885.71ms │
│ Total Time (groupjoin-eager-strategy)   │ 10939.81ms │
│ Average Time (HEAD)                     │   111.08ms │
│ Average Time (groupjoin-eager-strategy) │   111.63ms │
│ Queries Faster                          │          3 │
│ Queries Slower                          │          1 │
│ Queries with No Change                  │         94 │
│ Queries with Failure                    │          1 │
└─────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	55.0s
Peak memory	6.9 GiB
Avg memory	6.1 GiB
CPU user	251.9s
CPU sys	6.3s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	55.0s
Peak memory	6.9 GiB
Avg memory	6.2 GiB
CPU user	253.0s
CPU sys	6.3s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-05-07T14:24:23Z

🤖 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 groupjoin-eager-strategy
--------------------
Benchmark tpch_sf10.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                               HEAD ┃           groupjoin-eager-strategy ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │  338.65 / 340.66 ±1.78 / 343.87 ms │  339.37 / 340.47 ±0.85 / 341.69 ms │     no change │
│ QQuery 2  │  132.05 / 136.87 ±4.02 / 143.50 ms │  130.43 / 134.21 ±2.92 / 137.53 ms │     no change │
│ QQuery 3  │  286.84 / 289.73 ±1.98 / 293.01 ms │  287.12 / 290.32 ±2.24 / 293.02 ms │     no change │
│ QQuery 4  │  138.41 / 142.05 ±1.99 / 144.15 ms │  140.16 / 143.49 ±3.19 / 149.31 ms │     no change │
│ QQuery 5  │  424.90 / 431.37 ±3.96 / 435.88 ms │  424.02 / 431.41 ±6.37 / 440.03 ms │     no change │
│ QQuery 6  │  138.05 / 139.57 ±1.91 / 143.32 ms │  139.13 / 141.74 ±3.49 / 148.46 ms │     no change │
│ QQuery 7  │  556.81 / 565.33 ±7.72 / 578.96 ms │  568.42 / 570.22 ±1.56 / 572.49 ms │     no change │
│ QQuery 8  │  452.03 / 460.52 ±6.27 / 469.38 ms │  457.95 / 463.74 ±4.52 / 471.84 ms │     no change │
│ QQuery 9  │  658.09 / 669.69 ±7.87 / 677.08 ms │  653.55 / 663.84 ±7.22 / 675.65 ms │     no change │
│ QQuery 10 │  337.82 / 350.62 ±9.05 / 361.50 ms │  354.73 / 357.96 ±5.23 / 368.35 ms │     no change │
│ QQuery 11 │  101.34 / 107.47 ±3.44 / 110.51 ms │  100.94 / 105.90 ±6.41 / 118.49 ms │     no change │
│ QQuery 12 │  202.68 / 208.36 ±6.11 / 216.58 ms │  202.18 / 205.63 ±4.80 / 214.91 ms │     no change │
│ QQuery 13 │  318.65 / 328.94 ±6.96 / 337.68 ms │  274.45 / 280.78 ±5.98 / 290.93 ms │ +1.17x faster │
│ QQuery 14 │  192.73 / 194.32 ±1.64 / 197.27 ms │  189.43 / 192.60 ±2.36 / 196.62 ms │     no change │
│ QQuery 15 │  350.05 / 352.62 ±1.78 / 354.98 ms │  346.09 / 348.61 ±2.51 / 352.63 ms │     no change │
│ QQuery 16 │     79.05 / 81.16 ±1.70 / 83.53 ms │     79.33 / 81.68 ±1.66 / 84.15 ms │     no change │
│ QQuery 17 │  775.75 / 786.90 ±7.07 / 794.69 ms │  768.40 / 780.45 ±9.10 / 794.73 ms │     no change │
│ QQuery 18 │  805.07 / 817.93 ±7.87 / 825.54 ms │ 800.88 / 850.20 ±72.80 / 994.69 ms │     no change │
│ QQuery 19 │ 280.62 / 292.51 ±12.44 / 310.83 ms │ 278.64 / 294.84 ±15.73 / 314.81 ms │     no change │
│ QQuery 20 │  315.55 / 325.47 ±7.21 / 334.84 ms │ 311.59 / 330.95 ±10.06 / 338.29 ms │     no change │
│ QQuery 21 │  795.83 / 807.79 ±8.29 / 821.35 ms │  800.19 / 812.21 ±6.45 / 819.55 ms │     no change │
│ QQuery 22 │     69.63 / 71.80 ±1.57 / 73.84 ms │     70.09 / 71.52 ±1.13 / 72.95 ms │     no change │
└───────────┴────────────────────────────────────┴────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                       ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                       │ 7901.68ms │
│ Total Time (groupjoin-eager-strategy)   │ 7892.77ms │
│ Average Time (HEAD)                     │  359.17ms │
│ Average Time (groupjoin-eager-strategy) │  358.76ms │
│ Queries Faster                          │         1 │
│ Queries Slower                          │         0 │
│ Queries with No Change                  │        21 │
│ Queries with Failure                    │         0 │
└─────────────────────────────────────────┴───────────┘

Resource Usage

tpch10 — base (merge-base)

Metric	Value
Wall time	40.0s
Peak memory	12.4 GiB
Avg memory	8.6 GiB
CPU user	422.4s
CPU sys	23.6s
Peak spill	0 B

tpch10 — branch

Metric	Value
Wall time	40.0s
Peak memory	11.9 GiB
Avg memory	8.6 GiB
CPU user	420.8s
CPU sys	23.4s
Peak spill	0 B

File an issue against this benchmark runner

nathanb9 added 4 commits May 3, 2026 04:30

Add group join physical optimizer

5fe7219

Fix group join optimizer CI coverage

b35fd03

Fix group join CI failures

18dafff

Add eager group join strategy

1f38ecb

github-actions Bot added optimizer Optimizer rules core Core DataFusion crate sqllogictest SQL Logic Tests (.slt) proto Related to proto crate physical-plan Changes to the physical-plan crate labels May 7, 2026

github-actions Bot added the auto detected api change Auto detected API change label May 7, 2026

Implement eager group join execution

30075d5

nathanb9 mentioned this pull request May 7, 2026

Add group join physical optimizer #21995

Draft

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add eager group join strategy#22058

Add eager group join strategy#22058
nathanb9 wants to merge 5 commits intoapache:mainfrom
nathanb9:groupjoin-eager-strategy

nathanb9 commented May 7, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented May 7, 2026 •

edited

Loading

Uh oh!

nathanb9 commented May 7, 2026

Uh oh!

Dandandan commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

nathanb9 commented May 7, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented May 7, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

nathanb9 commented May 7, 2026

Uh oh!

Dandandan commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

adriangbot commented May 7, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

nathanb9 commented May 7, 2026 •

edited

Loading

github-actions Bot commented May 7, 2026 •

edited

Loading