Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangbot commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4260972838 🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21666#issuecomment-4260950110) **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4260950110-1371-8wgd5 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 codex/hash-join-empty-partition-reporting (0584854263a02d0ea2c0cd78cc25d7b2cf5b1e1f) to 5c653be (merge-base) [diff](https://github.com/apache/datafusion/compare/5c653bee5da64003915f6dfeb3da15759b091a8d..0584854263a02d0ea2c0cd78cc25d7b2cf5b1e1f) using: tpcds Results will be posted here when complete --- [File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangb commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4260950110 run benchmark tpcds -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangb commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4260535615 Yes. And I think this was another bandaid. But it's closer to the root cause than previous attempts. This has to do with cancellation when multiple joins are involved. TLDR I think what is happening is when you have multiple joins you end up with a tree of operators. One of the joins up higher in the tree hits the new optimization and aborts work, dropping tasks that would have polled downstream joins. But not the downstream join is stuck waiting for all of it's partition tasks to finish even though they never will. I think we were all operating under the assumption that the issue was within a single join operator but really it's an issue any time an upstream operator cancels on a join. I think the real solution is to track when a join build partition task gets dropped and report that to the dynamic filter building so that it doesn't wait for that partition to report. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
Omega359 commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4260406388 query 72 took a bit of a hit here. Query72 ```sql select i_item_desc ,w_warehouse_name ,d1.d_week_seq ,sum(case when p_promo_sk is null then 1 else 0 end) no_promo ,sum(case when p_promo_sk is not null then 1 else 0 end) promo ,count(*) total_cnt from catalog_sales join inventory on (cs_item_sk = inv_item_sk) join warehouse on (w_warehouse_sk=inv_warehouse_sk) join item on (i_item_sk = cs_item_sk) join customer_demographics on (cs_bill_cdemo_sk = cd_demo_sk) join household_demographics on (cs_bill_hdemo_sk = hd_demo_sk) join date_dim d1 on (cs_sold_date_sk = d1.d_date_sk) join date_dim d2 on (inv_date_sk = d2.d_date_sk) join date_dim d3 on (cs_ship_date_sk = d3.d_date_sk) left outer join promotion on (cs_promo_sk=p_promo_sk) left outer join catalog_returns on (cr_item_sk = cs_item_sk and cr_order_number = cs_order_number) where d1.d_week_seq = d2.d_week_seq and inv_quantity_on_hand < cs_quantity and d3.d_date > (d1.d_date + INTERVAL '5 days') and hd_buy_potential = '>1' and d1.d_year = 1999 and cd_marital_status = 'D' group by i_item_desc,w_warehouse_name,d1.d_week_seq order by total_cnt desc, i_item_desc, w_warehouse_name, d_week_seq limit 100; ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangbot commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4260230574 🤖 Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/21666#issuecomment-4260131019) **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 codex_hash-join-empty-partition-reporting Benchmark tpcds_sf1.json ┏━━━┳━━┳━━━┳━━━┓ ┃ Query ┃ HEAD ┃ codex_hash-join-empty-partition-reporting ┃Change ┃ ┡━━━╇━━╇━━━╇━━━┩ │ QQuery 1 │ 6.55 / 7.03 ±0.79 / 8.60 ms │ 6.62 / 7.11 ±0.87 / 8.84 ms │ no change │ │ QQuery 2 │143.53 / 143.94 ±0.32 / 144.23 ms │ 142.69 / 144.27 ±0.86 / 145.12 ms │ no change │ │ QQuery 3 │113.45 / 113.89 ±0.36 / 114.35 ms │ 113.86 / 114.43 ±0.60 / 115.53 ms │ no change │ │ QQuery 4 │1334.29 / 1362.57 ±21.81 / 1394.94 ms │ 1360.31 / 1390.10 ±16.45 / 1405.65 ms │ no change │ │ QQuery 5 │170.59 / 172.99 ±1.76 / 175.28 ms │ 172.45 / 174.47 ±1.77 / 177.23 ms │ no change │ │ QQuery 6 │ 826.41 / 876.05 ±36.80 / 938.37 ms │862.48 / 884.01 ±15.84 / 909.86 ms │ no change │ │ QQuery 7 │339.57 / 344.15 ±2.45 / 346.94 ms │ 340.62 / 345.25 ±2.84 / 348.46 ms │ no change │ │ QQuery 8 │115.88 / 116.71 ±0.58 / 117.66 ms │ 116.03 / 116.77 ±0.46 / 117.50 ms │ no change │ │ QQuery 9 │100.96 / 103.05 ±2.10 / 106.73 ms │ 100.29 / 106.99 ±5.92 / 114.42 ms │ no change │ │ QQuery 10 │106.11 / 106.48 ±0.35 / 107.03 ms │ 107.15 / 108.84 ±1.32 / 110.87 ms │ no change │ │ QQuery 11 │930.37 / 943.58 ±8.27 / 952.46 ms │ 956.74 / 960.95 ±3.40 / 966.02 ms │ no change │ │ QQuery 12 │ 44.35 / 46.11 ±1.12 / 47.41 ms │45.31 / 46.13 ±0.70 / 47.37 ms │ no change │ │ QQuery 13 │399.86 / 401.27 ±1.29 / 402.88 ms │ 400.47 / 403.82 ±2.05 / 406.38 ms │ no change │ │ QQuery 14 │ 992.83 / 1007.54 ±10.23 / 1022.21 ms │991.24 / 996.86 ±4.77 / 1002.00 ms │ no change │
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangbot commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4260142881 🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21666#issuecomment-4260131019) **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4260131019-1367-smhnv 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 codex/hash-join-empty-partition-reporting (d17d5e46a45d6fa62d3b2c7a2959cba662fb0e01) to 5c653be (merge-base) [diff](https://github.com/apache/datafusion/compare/5c653bee5da64003915f6dfeb3da15759b091a8d..d17d5e46a45d6fa62d3b2c7a2959cba662fb0e01) using: tpcds Results will be posted here when complete --- [File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangb commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4260131019 run benchmark tpcds -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangbot commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4259990887 🤖 Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/21666#issuecomment-4259875560) **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 codex_hash-join-empty-partition-reporting Benchmark tpcds_sf1.json ┏━━━┳━━┳━━━┳━━━┓ ┃ Query ┃ HEAD ┃ codex_hash-join-empty-partition-reporting ┃Change ┃ ┡━━━╇━━╇━━━╇━━━┩ │ QQuery 1 │ 6.98 / 7.37 ±0.69 / 8.75 ms │ 7.09 / 7.54 ±0.69 / 8.90 ms │ no change │ │ QQuery 2 │145.73 / 146.31 ±0.32 / 146.60 ms │ 146.27 / 147.38 ±0.76 / 148.07 ms │ no change │ │ QQuery 3 │114.06 / 114.75 ±0.58 / 115.63 ms │ 115.04 / 116.01 ±0.69 / 116.89 ms │ no change │ │ QQuery 4 │1410.96 / 1432.76 ±11.84 / 1445.23 ms │ 1421.36 / 1449.73 ±18.93 / 1470.63 ms │ no change │ │ QQuery 5 │174.22 / 175.39 ±1.01 / 177.11 ms │ 175.06 / 175.90 ±0.69 / 176.96 ms │ no change │ │ QQuery 6 │ 845.04 / 889.68 ±26.52 / 915.48 ms │857.43 / 886.16 ±21.55 / 910.23 ms │ no change │ │ QQuery 7 │348.38 / 350.63 ±2.63 / 354.78 ms │ 348.24 / 350.59 ±1.51 / 352.86 ms │ no change │ │ QQuery 8 │117.72 / 118.21 ±0.41 / 118.88 ms │ 117.47 / 118.30 ±0.76 / 119.43 ms │ no change │ │ QQuery 9 │102.35 / 105.91 ±2.84 / 109.15 ms │102.56 / 111.85 ±11.17 / 133.78 ms │ 1.06x slower │ │ QQuery 10 │107.24 / 108.52 ±0.75 / 109.28 ms │ 109.81 / 111.12 ±0.86 / 112.11 ms │ no change │ │ QQuery 11 │ 1007.00 / 1021.26 ±9.23 / 1034.34 ms │ 1007.07 / 1024.01 ±8.78 / 1031.00 ms │ no change │ │ QQuery 12 │ 45.77 / 49.05 ±1.75 / 50.94 ms │46.99 / 48.84 ±1.52 / 50.68 ms │ no change │ │ QQuery 13 │413.26 / 417.25 ±5.89 / 428.86 ms │ 408.54 / 411.67 ±3.27 / 417.80 ms │ no change │ │ QQuery 14 │ 1009.07 / 1017.56 ±7.45 / 1029.43 ms │ 999.17 / 1018.18 ±11.21 / 1031.36 ms │ no change │
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangbot commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4259989300 🤖 Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/21666#issuecomment-4259875560) **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 codex_hash-join-empty-partition-reporting Benchmark clickbench_partitioned.json ┏━━━┳┳━━━┳━━━┓ ┃ Query ┃ HEAD ┃ codex_hash-join-empty-partition-reporting ┃Change ┃ ┡━━━╇╇━━━╇━━━┩ │ QQuery 0 │ 1.19 / 4.37 ±6.28 / 16.93 ms │ 1.19 / 4.46 ±6.37 / 17.19 ms │ no change │ │ QQuery 1 │ 14.88 / 15.01 ±0.11 / 15.17 ms │15.14 / 15.45 ±0.17 / 15.63 ms │ no change │ │ QQuery 2 │ 43.43 / 43.74 ±0.26 / 44.21 ms │43.98 / 44.39 ±0.30 / 44.90 ms │ no change │ │ QQuery 3 │ 41.57 / 44.12 ±1.69 / 46.34 ms │42.77 / 44.97 ±1.35 / 46.94 ms │ no change │ │ QQuery 4 │ 282.16 / 293.79 ±9.62 / 307.92 ms │ 285.44 / 296.97 ±7.26 / 308.10 ms │ no change │ │ QQuery 5 │ 337.41 / 342.71 ±2.83 / 345.05 ms │ 352.81 / 358.06 ±4.34 / 364.86 ms │ no change │ │ QQuery 6 │4.98 / 5.41 ±0.22 / 5.55 ms │ 5.30 / 9.08 ±3.72 / 15.49 ms │ 1.68x slower │ │ QQuery 7 │ 16.78 / 17.38 ±0.47 / 18.15 ms │17.51 / 17.64 ±0.10 / 17.76 ms │ no change │ │ QQuery 8 │ 415.76 / 424.23 ±8.58 / 440.27 ms │410.23 / 429.54 ±12.05 / 446.87 ms │ no change │ │ QQuery 9 │ 648.84 / 658.48 ±9.24 / 671.44 ms │ 631.88 / 643.91 ±7.25 / 654.13 ms │ no change │ │ QQuery 10 │ 92.78 / 95.81 ±2.44 / 99.64 ms │92.15 / 94.38 ±1.43 / 96.26 ms │ no change │ │ QQuery 11 │ 105.44 / 106.03 ±0.58 / 107.09 ms │ 105.57 / 107.75 ±2.37 / 111.84 ms │ no change │ │ QQuery 12 │ 342.54 / 352.42 ±9.61 / 370.28 ms │ 334.21 / 339.65 ±3.70 / 345.28 ms │ no change │ │ QQuery 13 │ 461.57 / 470.47 ±8.18 / 484.10 ms │448.28 / 464.82 ±10.82 / 476.52 ms │ no change │ │ QQuery 14 │
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangbot commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4259891937 🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21666#issuecomment-4259875560) **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4259875560-1361-d7znb 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 codex/hash-join-empty-partition-reporting (d17d5e46a45d6fa62d3b2c7a2959cba662fb0e01) to 5c653be (merge-base) [diff](https://github.com/apache/datafusion/compare/5c653bee5da64003915f6dfeb3da15759b091a8d..d17d5e46a45d6fa62d3b2c7a2959cba662fb0e01) using: clickbench_partitioned Results will be posted here when complete --- [File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangbot commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4259892610 🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21666#issuecomment-4259875560) **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4259875560-1363-qrwrt 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 codex/hash-join-empty-partition-reporting (d17d5e46a45d6fa62d3b2c7a2959cba662fb0e01) to 5c653be (merge-base) [diff](https://github.com/apache/datafusion/compare/5c653bee5da64003915f6dfeb3da15759b091a8d..d17d5e46a45d6fa62d3b2c7a2959cba662fb0e01) using: tpch Results will be posted here when complete --- [File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangbot commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4259890973 🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21666#issuecomment-4259875560) **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4259875560-1362-nk27n 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 codex/hash-join-empty-partition-reporting (d17d5e46a45d6fa62d3b2c7a2959cba662fb0e01) to 5c653be (merge-base) [diff](https://github.com/apache/datafusion/compare/5c653bee5da64003915f6dfeb3da15759b091a8d..d17d5e46a45d6fa62d3b2c7a2959cba662fb0e01) using: tpcds Results will be posted here when complete --- [File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] Report empty build partitions for partitioned hash join filters [datafusion]
adriangb commented on PR #21666: URL: https://github.com/apache/datafusion/pull/21666#issuecomment-4259875560 run benchmarks -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
