datafusion
datafusion copied to clipboard
apache
feat: Support `u32` indices for `HashJoinExec`
Which issue does this PR close?
- Closes #16179 .
Rationale for this change
We can use u32 indices instead of u64 indices when there are less than u32::MAX rows when building the hashmap. This acts as a memory optimization
What changes are included in this PR?
During HashJoinExec we construct the JoinLeftData with a Box<dyn JoinHashMapType> choosing between a u32 indice or u64 JoinHashMap.
I changed the JoinHashMapType to hold the update_from_iter, get_matched_indice, and get_matched_indices_with_limit_offset and split the JoinHashMap into JoinHashMapU32 and JoinHashMapU64.
I deliberately did not try to expose a generic in the trait, nor did I try to create a generic on the JoinHashMap struct as doing so would lead to many problems with having to call preceding functions with a generic; doing so is not possible since we are determining the JoinHashMapType during runtime.
Are these changes tested?
Yes I added a test for checking the hashmap created using u32 indices.
🤖 ./gh_compare_branch.sh Benchmark Script Running
Linux aal-dev 6.11.0-1015-gcp #15~24.04.1-Ubuntu SMP Thu Apr 24 20:41:05 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing support-u32-hashmap (732fa2187396749b4abc40b034b4d60db396b403) to 056f5462bbc3de620f55e793343bc3a2276bd3d7 diff
Benchmarks: tpch_mem clickbench_partitioned clickbench_extended
Results will be posted here when complete
🤖: Benchmark completed
Details
Comparing HEAD and support-u32-hashmap
--------------------
Benchmark clickbench_extended.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0 │ 1891.38 ms │ 1886.58 ms │ no change │
│ QQuery 1 │ 697.56 ms │ 704.74 ms │ no change │
│ QQuery 2 │ 1381.91 ms │ 1361.41 ms │ no change │
│ QQuery 3 │ 669.87 ms │ 634.92 ms │ +1.06x faster │
│ QQuery 4 │ 1327.15 ms │ 1339.32 ms │ no change │
│ QQuery 5 │ 14912.55 ms │ 14887.37 ms │ no change │
│ QQuery 6 │ 2044.07 ms │ 2059.57 ms │ no change │
│ QQuery 7 │ 1784.47 ms │ 1873.16 ms │ no change │
│ QQuery 8 │ 797.99 ms │ 802.46 ms │ no change │
└──────────────┴─────────────┴─────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD) │ 25506.95ms │
│ Total Time (support-u32-hashmap) │ 25549.53ms │
│ Average Time (HEAD) │ 2834.11ms │
│ Average Time (support-u32-hashmap) │ 2838.84ms │
│ Queries Faster │ 1 │
│ Queries Slower │ 0 │
│ Queries with No Change │ 8 │
│ Queries with Failure │ 0 │
└────────────────────────────────────┴────────────┘
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 0 │ 15.40 ms │ 15.56 ms │ no change │
│ QQuery 1 │ 33.85 ms │ 33.67 ms │ no change │
│ QQuery 2 │ 81.80 ms │ 80.27 ms │ no change │
│ QQuery 3 │ 94.51 ms │ 98.91 ms │ no change │
│ QQuery 4 │ 580.34 ms │ 587.97 ms │ no change │
│ QQuery 5 │ 818.51 ms │ 852.27 ms │ no change │
│ QQuery 6 │ 22.18 ms │ 23.65 ms │ 1.07x slower │
│ QQuery 7 │ 36.35 ms │ 36.60 ms │ no change │
│ QQuery 8 │ 840.07 ms │ 846.70 ms │ no change │
│ QQuery 9 │ 1163.06 ms │ 1126.59 ms │ no change │
│ QQuery 10 │ 256.44 ms │ 253.13 ms │ no change │
│ QQuery 11 │ 284.27 ms │ 275.93 ms │ no change │
│ QQuery 12 │ 855.55 ms │ 892.86 ms │ no change │
│ QQuery 13 │ 1240.05 ms │ 1256.85 ms │ no change │
│ QQuery 14 │ 788.52 ms │ 799.55 ms │ no change │
│ QQuery 15 │ 756.94 ms │ 759.38 ms │ no change │
│ QQuery 16 │ 1588.12 ms │ 1573.61 ms │ no change │
│ QQuery 17 │ 1596.56 ms │ 1595.53 ms │ no change │
│ QQuery 18 │ 2844.62 ms │ 2904.45 ms │ no change │
│ QQuery 19 │ 82.55 ms │ 85.96 ms │ no change │
│ QQuery 20 │ 1133.70 ms │ 1187.01 ms │ no change │
│ QQuery 21 │ 1273.35 ms │ 1310.45 ms │ no change │
│ QQuery 22 │ 2112.94 ms │ 2171.74 ms │ no change │
│ QQuery 23 │ 7356.55 ms │ 7416.88 ms │ no change │
│ QQuery 24 │ 427.69 ms │ 444.19 ms │ no change │
│ QQuery 25 │ 299.39 ms │ 302.95 ms │ no change │
│ QQuery 26 │ 433.22 ms │ 449.51 ms │ no change │
│ QQuery 27 │ 1555.11 ms │ 1553.18 ms │ no change │
│ QQuery 28 │ 11668.95 ms │ 11887.97 ms │ no change │
│ QQuery 29 │ 519.95 ms │ 508.95 ms │ no change │
│ QQuery 30 │ 768.44 ms │ 778.51 ms │ no change │
│ QQuery 31 │ 799.04 ms │ 818.27 ms │ no change │
│ QQuery 32 │ 2409.37 ms │ 2374.14 ms │ no change │
│ QQuery 33 │ 3138.93 ms │ 3153.14 ms │ no change │
│ QQuery 34 │ 3162.98 ms │ 3160.19 ms │ no change │
│ QQuery 35 │ 1212.81 ms │ 1231.29 ms │ no change │
│ QQuery 36 │ 123.25 ms │ 125.52 ms │ no change │
│ QQuery 37 │ 57.19 ms │ 57.98 ms │ no change │
│ QQuery 38 │ 127.02 ms │ 124.89 ms │ no change │
│ QQuery 39 │ 195.28 ms │ 197.94 ms │ no change │
│ QQuery 40 │ 47.18 ms │ 48.68 ms │ no change │
│ QQuery 41 │ 45.15 ms │ 42.97 ms │ no change │
│ QQuery 42 │ 38.79 ms │ 39.72 ms │ no change │
└──────────────┴─────────────┴─────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD) │ 52886.01ms │
│ Total Time (support-u32-hashmap) │ 53485.49ms │
│ Average Time (HEAD) │ 1229.91ms │
│ Average Time (support-u32-hashmap) │ 1243.85ms │
│ Queries Faster │ 0 │
│ Queries Slower │ 1 │
│ Queries with No Change │ 42 │
│ Queries with Failure │ 0 │
└────────────────────────────────────┴────────────┘
--------------------
Benchmark tpch_mem_sf1.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1 │ 100.68 ms │ 99.40 ms │ no change │
│ QQuery 2 │ 21.39 ms │ 21.69 ms │ no change │
│ QQuery 3 │ 32.64 ms │ 32.28 ms │ no change │
│ QQuery 4 │ 18.13 ms │ 18.58 ms │ no change │
│ QQuery 5 │ 48.89 ms │ 49.42 ms │ no change │
│ QQuery 6 │ 11.87 ms │ 11.86 ms │ no change │
│ QQuery 7 │ 87.37 ms │ 83.45 ms │ no change │
│ QQuery 8 │ 23.84 ms │ 23.90 ms │ no change │
│ QQuery 9 │ 53.64 ms │ 53.53 ms │ no change │
│ QQuery 10 │ 42.80 ms │ 42.86 ms │ no change │
│ QQuery 11 │ 11.21 ms │ 11.22 ms │ no change │
│ QQuery 12 │ 34.76 ms │ 34.72 ms │ no change │
│ QQuery 13 │ 26.15 ms │ 26.24 ms │ no change │
│ QQuery 14 │ 9.82 ms │ 9.88 ms │ no change │
│ QQuery 15 │ 19.41 ms │ 19.59 ms │ no change │
│ QQuery 16 │ 18.81 ms │ 18.85 ms │ no change │
│ QQuery 17 │ 96.08 ms │ 94.52 ms │ no change │
│ QQuery 18 │ 193.90 ms │ 189.03 ms │ no change │
│ QQuery 19 │ 25.54 ms │ 25.12 ms │ no change │
│ QQuery 20 │ 33.31 ms │ 31.69 ms │ no change │
│ QQuery 21 │ 148.25 ms │ 146.30 ms │ no change │
│ QQuery 22 │ 15.18 ms │ 15.31 ms │ no change │
└──────────────┴───────────┴─────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD) │ 1073.66ms │
│ Total Time (support-u32-hashmap) │ 1059.45ms │
│ Average Time (HEAD) │ 48.80ms │
│ Average Time (support-u32-hashmap) │ 48.16ms │
│ Queries Faster │ 0 │
│ Queries Slower │ 0 │
│ Queries with No Change │ 22 │
│ Queries with Failure │ 0 │
└────────────────────────────────────┴───────────┘
@Dandandan I can see if I can run some benchmarks.
@alamb this should be good to go, i'll see if I can implement some much needed hash join spilling after this gets merged 🚀
🤖 ./gh_compare_branch.sh Benchmark Script Running
Linux aal-dev 6.11.0-1016-gcp #16~24.04.1-Ubuntu SMP Wed May 28 02:40:52 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing support-u32-hashmap (6c1543d549bdd90603b91e8c4644b5545b8fe44e) to ebb8e9591322ee0a6fbc42cd43ce60f0e20981ba diff using: tpch_mem clickbench_partitioned clickbench_extended
Results will be posted here when complete
🤖: Benchmark completed
Details
Comparing HEAD and support-u32-hashmap
--------------------
Benchmark clickbench_extended.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0 │ 2104.73 ms │ 1846.85 ms │ +1.14x faster │
│ QQuery 1 │ 694.10 ms │ 813.41 ms │ 1.17x slower │
│ QQuery 2 │ 1312.83 ms │ 1507.45 ms │ 1.15x slower │
│ QQuery 3 │ 691.45 ms │ 678.94 ms │ no change │
│ QQuery 4 │ 1373.17 ms │ 1348.65 ms │ no change │
│ QQuery 5 │ 15124.71 ms │ 15473.54 ms │ no change │
│ QQuery 6 │ 2047.34 ms │ 2060.53 ms │ no change │
│ QQuery 7 │ 1873.73 ms │ 1838.61 ms │ no change │
└──────────────┴─────────────┴─────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD) │ 25222.05ms │
│ Total Time (support-u32-hashmap) │ 25567.98ms │
│ Average Time (HEAD) │ 3152.76ms │
│ Average Time (support-u32-hashmap) │ 3196.00ms │
│ Queries Faster │ 1 │
│ Queries Slower │ 2 │
│ Queries with No Change │ 5 │
│ Queries with Failure │ 0 │
└────────────────────────────────────┴────────────┘
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0 │ 2.43 ms │ 2.18 ms │ +1.11x faster │
│ QQuery 1 │ 35.82 ms │ 34.78 ms │ no change │
│ QQuery 2 │ 82.71 ms │ 82.00 ms │ no change │
│ QQuery 3 │ 97.18 ms │ 97.71 ms │ no change │
│ QQuery 4 │ 649.57 ms │ 589.40 ms │ +1.10x faster │
│ QQuery 5 │ 880.40 ms │ 885.32 ms │ no change │
│ QQuery 6 │ 2.31 ms │ 2.19 ms │ +1.06x faster │
│ QQuery 7 │ 38.57 ms │ 39.24 ms │ no change │
│ QQuery 8 │ 875.15 ms │ 871.19 ms │ no change │
│ QQuery 9 │ 1204.58 ms │ 1197.94 ms │ no change │
│ QQuery 10 │ 256.41 ms │ 264.73 ms │ no change │
│ QQuery 11 │ 295.15 ms │ 294.16 ms │ no change │
│ QQuery 12 │ 876.87 ms │ 898.49 ms │ no change │
│ QQuery 13 │ 1253.67 ms │ 1260.00 ms │ no change │
│ QQuery 14 │ 805.60 ms │ 822.48 ms │ no change │
│ QQuery 15 │ 775.34 ms │ 792.68 ms │ no change │
│ QQuery 16 │ 1582.94 ms │ 1617.44 ms │ no change │
│ QQuery 17 │ 1575.57 ms │ 1620.19 ms │ no change │
│ QQuery 18 │ 2812.98 ms │ 2877.62 ms │ no change │
│ QQuery 19 │ 86.50 ms │ 86.36 ms │ no change │
│ QQuery 20 │ 1147.33 ms │ 1142.38 ms │ no change │
│ QQuery 21 │ 1304.30 ms │ 1299.10 ms │ no change │
│ QQuery 22 │ 2109.41 ms │ 2165.23 ms │ no change │
│ QQuery 23 │ 7382.16 ms │ 7578.04 ms │ no change │
│ QQuery 24 │ 433.78 ms │ 449.56 ms │ no change │
│ QQuery 25 │ 305.40 ms │ 308.23 ms │ no change │
│ QQuery 26 │ 428.78 ms │ 444.90 ms │ no change │
│ QQuery 27 │ 1537.21 ms │ 1565.57 ms │ no change │
│ QQuery 28 │ 12883.98 ms │ 11966.19 ms │ +1.08x faster │
│ QQuery 29 │ 532.77 ms │ 513.35 ms │ no change │
│ QQuery 30 │ 765.17 ms │ 783.38 ms │ no change │
│ QQuery 31 │ 770.79 ms │ 808.37 ms │ no change │
│ QQuery 32 │ 2367.00 ms │ 2361.17 ms │ no change │
│ QQuery 33 │ 3106.32 ms │ 3186.17 ms │ no change │
│ QQuery 34 │ 3283.80 ms │ 3176.69 ms │ no change │
│ QQuery 35 │ 1228.46 ms │ 1254.86 ms │ no change │
│ QQuery 36 │ 122.51 ms │ 122.45 ms │ no change │
│ QQuery 37 │ 54.42 ms │ 52.77 ms │ no change │
│ QQuery 38 │ 118.02 ms │ 118.88 ms │ no change │
│ QQuery 39 │ 196.50 ms │ 194.35 ms │ no change │
│ QQuery 40 │ 40.97 ms │ 40.46 ms │ no change │
│ QQuery 41 │ 39.08 ms │ 38.28 ms │ no change │
│ QQuery 42 │ 33.07 ms │ 33.30 ms │ no change │
└──────────────┴─────────────┴─────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD) │ 54381.00ms │
│ Total Time (support-u32-hashmap) │ 53939.79ms │
│ Average Time (HEAD) │ 1264.67ms │
│ Average Time (support-u32-hashmap) │ 1254.41ms │
│ Queries Faster │ 4 │
│ Queries Slower │ 0 │
│ Queries with No Change │ 39 │
│ Queries with Failure │ 0 │
└────────────────────────────────────┴────────────┘
--------------------
Benchmark tpch_mem_sf1.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1 │ 98.54 ms │ 97.07 ms │ no change │
│ QQuery 2 │ 20.04 ms │ 20.71 ms │ no change │
│ QQuery 3 │ 32.35 ms │ 31.90 ms │ no change │
│ QQuery 4 │ 18.71 ms │ 18.54 ms │ no change │
│ QQuery 5 │ 50.28 ms │ 50.68 ms │ no change │
│ QQuery 6 │ 11.77 ms │ 11.75 ms │ no change │
│ QQuery 7 │ 93.46 ms │ 88.13 ms │ +1.06x faster │
│ QQuery 8 │ 25.49 ms │ 25.07 ms │ no change │
│ QQuery 9 │ 54.54 ms │ 53.28 ms │ no change │
│ QQuery 10 │ 43.48 ms │ 42.65 ms │ no change │
│ QQuery 11 │ 11.37 ms │ 11.35 ms │ no change │
│ QQuery 12 │ 36.02 ms │ 34.57 ms │ no change │
│ QQuery 13 │ 26.86 ms │ 26.67 ms │ no change │
│ QQuery 14 │ 9.77 ms │ 9.57 ms │ no change │
│ QQuery 15 │ 19.01 ms │ 18.98 ms │ no change │
│ QQuery 16 │ 18.38 ms │ 18.23 ms │ no change │
│ QQuery 17 │ 96.47 ms │ 97.63 ms │ no change │
│ QQuery 18 │ 185.86 ms │ 201.01 ms │ 1.08x slower │
│ QQuery 19 │ 24.92 ms │ 24.58 ms │ no change │
│ QQuery 20 │ 31.25 ms │ 32.14 ms │ no change │
│ QQuery 21 │ 146.24 ms │ 143.79 ms │ no change │
│ QQuery 22 │ 14.79 ms │ 14.93 ms │ no change │
└──────────────┴───────────┴─────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD) │ 1069.58ms │
│ Total Time (support-u32-hashmap) │ 1073.22ms │
│ Average Time (HEAD) │ 48.62ms │
│ Average Time (support-u32-hashmap) │ 48.78ms │
│ Queries Faster │ 1 │
│ Queries Slower │ 1 │
│ Queries with No Change │ 20 │
│ Queries with Failure │ 0 │
└────────────────────────────────────┴───────────┘
🤖 ./gh_compare_branch.sh Benchmark Script Running
Linux aal-dev 6.11.0-1016-gcp #16~24.04.1-Ubuntu SMP Wed May 28 02:40:52 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing support-u32-hashmap (6c1543d549bdd90603b91e8c4644b5545b8fe44e) to ebb8e9591322ee0a6fbc42cd43ce60f0e20981ba diff using: tpch_mem10
Results will be posted here when complete
🤖: Benchmark completed
Details
Comparing HEAD and support-u32-hashmap
--------------------
Benchmark tpch_mem_sf10.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1 │ FAIL │ FAIL │ incomparable │
│ QQuery 2 │ FAIL │ FAIL │ incomparable │
│ QQuery 3 │ FAIL │ FAIL │ incomparable │
│ QQuery 4 │ FAIL │ FAIL │ incomparable │
│ QQuery 5 │ FAIL │ FAIL │ incomparable │
│ QQuery 6 │ FAIL │ FAIL │ incomparable │
│ QQuery 7 │ FAIL │ FAIL │ incomparable │
│ QQuery 8 │ FAIL │ FAIL │ incomparable │
│ QQuery 9 │ FAIL │ FAIL │ incomparable │
│ QQuery 10 │ FAIL │ FAIL │ incomparable │
│ QQuery 11 │ FAIL │ FAIL │ incomparable │
│ QQuery 12 │ FAIL │ FAIL │ incomparable │
│ QQuery 13 │ FAIL │ FAIL │ incomparable │
│ QQuery 14 │ FAIL │ FAIL │ incomparable │
│ QQuery 15 │ FAIL │ FAIL │ incomparable │
│ QQuery 16 │ FAIL │ FAIL │ incomparable │
│ QQuery 17 │ FAIL │ FAIL │ incomparable │
│ QQuery 18 │ FAIL │ FAIL │ incomparable │
│ QQuery 19 │ FAIL │ FAIL │ incomparable │
│ QQuery 20 │ FAIL │ FAIL │ incomparable │
│ QQuery 21 │ FAIL │ FAIL │ incomparable │
│ QQuery 22 │ FAIL │ FAIL │ incomparable │
└──────────────┴──────┴─────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩
│ Total Time (HEAD) │ 0.00ms │
│ Total Time (support-u32-hashmap) │ 0.00ms │
│ Average Time (HEAD) │ 0.00ms │
│ Average Time (support-u32-hashmap) │ 0.00ms │
│ Queries Faster │ 0 │
│ Queries Slower │ 0 │
│ Queries with No Change │ 0 │
│ Queries with Failure │ 22 │
└────────────────────────────────────┴────────┘
🤖 ./gh_compare_branch.sh Benchmark Script Running
Linux aal-dev 6.11.0-1016-gcp #16~24.04.1-Ubuntu SMP Wed May 28 02:40:52 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing support-u32-hashmap (6c1543d549bdd90603b91e8c4644b5545b8fe44e) to ebb8e9591322ee0a6fbc42cd43ce60f0e20981ba diff using: tpch_mem10
Results will be posted here when complete
🤖: Benchmark completed
Details
Comparing HEAD and support-u32-hashmap
--------------------
Benchmark tpch_mem_sf10.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1 │ FAIL │ FAIL │ incomparable │
│ QQuery 2 │ FAIL │ FAIL │ incomparable │
│ QQuery 3 │ FAIL │ FAIL │ incomparable │
│ QQuery 4 │ FAIL │ FAIL │ incomparable │
│ QQuery 5 │ FAIL │ FAIL │ incomparable │
│ QQuery 6 │ FAIL │ FAIL │ incomparable │
│ QQuery 7 │ FAIL │ FAIL │ incomparable │
│ QQuery 8 │ FAIL │ FAIL │ incomparable │
│ QQuery 9 │ FAIL │ FAIL │ incomparable │
│ QQuery 10 │ FAIL │ FAIL │ incomparable │
│ QQuery 11 │ FAIL │ FAIL │ incomparable │
│ QQuery 12 │ FAIL │ FAIL │ incomparable │
│ QQuery 13 │ FAIL │ FAIL │ incomparable │
│ QQuery 14 │ FAIL │ FAIL │ incomparable │
│ QQuery 15 │ FAIL │ FAIL │ incomparable │
│ QQuery 16 │ FAIL │ FAIL │ incomparable │
│ QQuery 17 │ FAIL │ FAIL │ incomparable │
│ QQuery 18 │ FAIL │ FAIL │ incomparable │
│ QQuery 19 │ FAIL │ FAIL │ incomparable │
│ QQuery 20 │ FAIL │ FAIL │ incomparable │
│ QQuery 21 │ FAIL │ FAIL │ incomparable │
│ QQuery 22 │ FAIL │ FAIL │ incomparable │
└──────────────┴──────┴─────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩
│ Total Time (HEAD) │ 0.00ms │
│ Total Time (support-u32-hashmap) │ 0.00ms │
│ Average Time (HEAD) │ 0.00ms │
│ Average Time (support-u32-hashmap) │ 0.00ms │
│ Queries Faster │ 0 │
│ Queries Slower │ 0 │
│ Queries with No Change │ 0 │
│ Queries with Failure │ 22 │
└────────────────────────────────────┴────────┘
🤖 ./gh_compare_branch.sh Benchmark Script Running
Linux aal-dev 6.11.0-1016-gcp #16~24.04.1-Ubuntu SMP Wed May 28 02:40:52 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing support-u32-hashmap (6c1543d549bdd90603b91e8c4644b5545b8fe44e) to ebb8e9591322ee0a6fbc42cd43ce60f0e20981ba diff using: tpch_mem10
Results will be posted here when complete
🤖 ./gh_compare_branch.sh Benchmark Script Running
Linux aal-dev 6.11.0-1016-gcp #16~24.04.1-Ubuntu SMP Wed May 28 02:40:52 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing support-u32-hashmap (6c1543d549bdd90603b91e8c4644b5545b8fe44e) to ebb8e9591322ee0a6fbc42cd43ce60f0e20981ba diff using: tpch_mem clickbench_partitioned clickbench_extended
Results will be posted here when complete
🤖: Benchmark completed
Details
Comparing HEAD and support-u32-hashmap
--------------------
Benchmark clickbench_extended.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 0 │ 1930.70 ms │ 1957.94 ms │ no change │
│ QQuery 1 │ 677.21 ms │ 798.33 ms │ 1.18x slower │
│ QQuery 2 │ 1357.45 ms │ 1481.82 ms │ 1.09x slower │
│ QQuery 3 │ 681.52 ms │ 678.49 ms │ no change │
│ QQuery 4 │ 1382.40 ms │ 1383.83 ms │ no change │
│ QQuery 5 │ 15028.67 ms │ 15023.36 ms │ no change │
│ QQuery 6 │ 2054.14 ms │ 2068.44 ms │ no change │
│ QQuery 7 │ 1933.85 ms │ 1881.38 ms │ no change │
└──────────────┴─────────────┴─────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD) │ 25045.95ms │
│ Total Time (support-u32-hashmap) │ 25273.59ms │
│ Average Time (HEAD) │ 3130.74ms │
│ Average Time (support-u32-hashmap) │ 3159.20ms │
│ Queries Faster │ 0 │
│ Queries Slower │ 2 │
│ Queries with No Change │ 6 │
│ Queries with Failure │ 0 │
└────────────────────────────────────┴────────────┘
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0 │ 2.21 ms │ 2.71 ms │ 1.22x slower │
│ QQuery 1 │ 33.32 ms │ 34.20 ms │ no change │
│ QQuery 2 │ 79.77 ms │ 80.70 ms │ no change │
│ QQuery 3 │ 99.32 ms │ 98.22 ms │ no change │
│ QQuery 4 │ 592.98 ms │ 588.77 ms │ no change │
│ QQuery 5 │ 827.02 ms │ 838.44 ms │ no change │
│ QQuery 6 │ 2.24 ms │ 2.25 ms │ no change │
│ QQuery 7 │ 38.09 ms │ 38.69 ms │ no change │
│ QQuery 8 │ 853.29 ms │ 853.08 ms │ no change │
│ QQuery 9 │ 1136.14 ms │ 1152.72 ms │ no change │
│ QQuery 10 │ 261.54 ms │ 261.42 ms │ no change │
│ QQuery 11 │ 285.91 ms │ 293.48 ms │ no change │
│ QQuery 12 │ 843.87 ms │ 862.33 ms │ no change │
│ QQuery 13 │ 1263.17 ms │ 1227.42 ms │ no change │
│ QQuery 14 │ 794.84 ms │ 811.95 ms │ no change │
│ QQuery 15 │ 758.54 ms │ 773.98 ms │ no change │
│ QQuery 16 │ 1625.44 ms │ 1593.33 ms │ no change │
│ QQuery 17 │ 1583.75 ms │ 1604.20 ms │ no change │
│ QQuery 18 │ 2820.29 ms │ 2840.30 ms │ no change │
│ QQuery 19 │ 86.91 ms │ 84.85 ms │ no change │
│ QQuery 20 │ 1149.21 ms │ 1149.69 ms │ no change │
│ QQuery 21 │ 1284.54 ms │ 1308.73 ms │ no change │
│ QQuery 22 │ 2113.86 ms │ 2149.07 ms │ no change │
│ QQuery 23 │ 7351.12 ms │ 7555.84 ms │ no change │
│ QQuery 24 │ 441.28 ms │ 445.38 ms │ no change │
│ QQuery 25 │ 298.98 ms │ 310.10 ms │ no change │
│ QQuery 26 │ 443.05 ms │ 434.69 ms │ no change │
│ QQuery 27 │ 1540.43 ms │ 1535.79 ms │ no change │
│ QQuery 28 │ 12749.97 ms │ 11922.38 ms │ +1.07x faster │
│ QQuery 29 │ 539.74 ms │ 515.15 ms │ no change │
│ QQuery 30 │ 772.49 ms │ 781.73 ms │ no change │
│ QQuery 31 │ 788.43 ms │ 797.62 ms │ no change │
│ QQuery 32 │ 2360.44 ms │ 2395.64 ms │ no change │
│ QQuery 33 │ 3139.22 ms │ 3142.80 ms │ no change │
│ QQuery 34 │ 3171.83 ms │ 3192.95 ms │ no change │
│ QQuery 35 │ 1223.89 ms │ 1247.48 ms │ no change │
│ QQuery 36 │ 119.12 ms │ 123.31 ms │ no change │
│ QQuery 37 │ 51.06 ms │ 50.90 ms │ no change │
│ QQuery 38 │ 117.58 ms │ 121.28 ms │ no change │
│ QQuery 39 │ 192.58 ms │ 197.01 ms │ no change │
│ QQuery 40 │ 43.25 ms │ 41.98 ms │ no change │
│ QQuery 41 │ 39.19 ms │ 38.33 ms │ no change │
│ QQuery 42 │ 32.36 ms │ 32.12 ms │ no change │
└──────────────┴─────────────┴─────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD) │ 53952.27ms │
│ Total Time (support-u32-hashmap) │ 53533.03ms │
│ Average Time (HEAD) │ 1254.70ms │
│ Average Time (support-u32-hashmap) │ 1244.95ms │
│ Queries Faster │ 1 │
│ Queries Slower │ 1 │
│ Queries with No Change │ 41 │
│ Queries with Failure │ 0 │
└────────────────────────────────────┴────────────┘
--------------------
Benchmark tpch_mem_sf1.json
--------------------
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ support-u32-hashmap ┃ Change ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1 │ 99.36 ms │ 97.70 ms │ no change │
│ QQuery 2 │ 21.37 ms │ 20.73 ms │ no change │
│ QQuery 3 │ 32.39 ms │ 32.45 ms │ no change │
│ QQuery 4 │ 18.84 ms │ 18.05 ms │ no change │
│ QQuery 5 │ 50.41 ms │ 50.26 ms │ no change │
│ QQuery 6 │ 11.88 ms │ 11.61 ms │ no change │
│ QQuery 7 │ 89.55 ms │ 84.48 ms │ +1.06x faster │
│ QQuery 8 │ 25.58 ms │ 23.78 ms │ +1.08x faster │
│ QQuery 9 │ 54.98 ms │ 53.07 ms │ no change │
│ QQuery 10 │ 42.29 ms │ 42.47 ms │ no change │
│ QQuery 11 │ 11.34 ms │ 11.28 ms │ no change │
│ QQuery 12 │ 35.44 ms │ 35.15 ms │ no change │
│ QQuery 13 │ 26.09 ms │ 25.54 ms │ no change │
│ QQuery 14 │ 9.81 ms │ 9.69 ms │ no change │
│ QQuery 15 │ 19.41 ms │ 18.82 ms │ no change │
│ QQuery 16 │ 17.99 ms │ 17.77 ms │ no change │
│ QQuery 17 │ 96.76 ms │ 93.99 ms │ no change │
│ QQuery 18 │ 190.99 ms │ 189.44 ms │ no change │
│ QQuery 19 │ 24.77 ms │ 24.99 ms │ no change │
│ QQuery 20 │ 31.07 ms │ 31.07 ms │ no change │
│ QQuery 21 │ 145.13 ms │ 145.34 ms │ no change │
│ QQuery 22 │ 14.74 ms │ 14.71 ms │ no change │
└──────────────┴───────────┴─────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD) │ 1070.19ms │
│ Total Time (support-u32-hashmap) │ 1052.40ms │
│ Average Time (HEAD) │ 48.65ms │
│ Average Time (support-u32-hashmap) │ 47.84ms │
│ Queries Faster │ 2 │
│ Queries Slower │ 0 │
│ Queries with No Change │ 20 │
│ Queries with Failure │ 0 │
└────────────────────────────────────┴───────────┘
TLDR this branch looks good from my performance perspective. Thank you @jonathanc-n and @Dandandan