Re: [PR] (WIP) Upgrade to arrow/parquet 55 [datafusion]

2025-04-10 Thread via GitHub
mbutrovich commented on PR #15466: URL: https://github.com/apache/datafusion/pull/15466#issuecomment-2794522838 Still seeing if this is just noise, but here are flame graphs for Q14 from my machine if anyone else wants to stare at them: This PR: ![pr](https://github.com/user-attac

Re: [PR] (WIP) Upgrade to arrow/parquet 55 [datafusion]

2025-04-10 Thread via GitHub
alamb commented on PR #15466: URL: https://github.com/apache/datafusion/pull/15466#issuecomment-2792347897 I tried briefly to reproduce the performance improvements reported above and it seems like I can: [q14.sql.txt](https://github.com/user-attachments/files/19683249/q14.sql.txt)

Re: [PR] (WIP) Upgrade to arrow/parquet 55 [datafusion]

2025-04-09 Thread via GitHub
Dandandan commented on PR #15466: URL: https://github.com/apache/datafusion/pull/15466#issuecomment-2791658055 I wonder if `alamb_test_upgrade_54` has the latest version of 54? Some performance improvements happened there as well (e.g. https://github.com/apache/arrow-rs/pull/7195/files shou

Re: [PR] (WIP) Upgrade to arrow/parquet 55 [datafusion]

2025-04-09 Thread via GitHub
alamb commented on PR #15466: URL: https://github.com/apache/datafusion/pull/15466#issuecomment-2789356373 🤖 `./gh_compare_branch.sh` [Benchmark Script](https://github.com/alamb/datafusion-benchmarking) Running Linux aal-dev 6.8.0-1016-gcp #18-Ubuntu SMP Fri Oct 4 22:16:29 UTC 2024 x86_

Re: [PR] (WIP) Upgrade to arrow/parquet 55 [datafusion]

2025-04-09 Thread via GitHub
alamb commented on PR #15466: URL: https://github.com/apache/datafusion/pull/15466#issuecomment-2789500250 🤖: Benchmark completed Details ``` Comparing HEAD and alamb_test_upgrade_54 Benchmark clickbench_extended.json

Re: [PR] (WIP) Upgrade to arrow/parquet 55 [datafusion]

2025-04-09 Thread via GitHub
alamb commented on PR #15466: URL: https://github.com/apache/datafusion/pull/15466#issuecomment-2789363141 🤖 `./gh_compare_branch.sh` [Benchmark Script](https://github.com/alamb/datafusion-benchmarking) Running Linux aal-dev 6.8.0-1016-gcp #18-Ubuntu SMP Fri Oct 4 22:16:29 UTC 2024 x86_

Re: [PR] (WIP) Upgrade to arrow/parquet 55 [datafusion]

2025-04-09 Thread via GitHub
alamb commented on PR #15466: URL: https://github.com/apache/datafusion/pull/15466#issuecomment-2789353839 🤖 `./gh_compare_branch.sh` [Benchmark Script](https://github.com/alamb/datafusion-benchmarking) Running Linux aal-dev 6.8.0-1016-gcp #18-Ubuntu SMP Fri Oct 4 22:16:29 UTC 2024 x86_

Re: [PR] (WIP) Upgrade to arrow/parquet 55 [datafusion]

2025-04-09 Thread via GitHub
alamb commented on PR #15466: URL: https://github.com/apache/datafusion/pull/15466#issuecomment-2789352434 🤖 `./gh_compare_branch.sh` [Benchmark Script](https://github.com/alamb/datafusion-benchmarking) Running Linux aal-dev 6.8.0-1016-gcp #18-Ubuntu SMP Fri Oct 4 22:16:29 UTC 2024 x86_

Re: [PR] (WIP) Upgrade to arrow/parquet 55 [datafusion]

2025-04-08 Thread via GitHub
alamb commented on PR #15466: URL: https://github.com/apache/datafusion/pull/15466#issuecomment-2787488417 🤖 `./gh_compare_branch.sh` [Benchmark Script](https://github.com/alamb/datafusion-benchmarking) Running Linux aal-dev 6.8.0-1016-gcp #18-Ubuntu SMP Fri Oct 4 22:16:29 UTC 2024 x86_

Re: [PR] (WIP) Upgrade to arrow/parquet 55 [datafusion]

2025-04-08 Thread via GitHub
alamb commented on code in PR #15466: URL: https://github.com/apache/datafusion/pull/15466#discussion_r2033373202 ## datafusion/functions-aggregate/benches/array_agg.rs: ## @@ -19,17 +19,23 @@ use std::sync::Arc; use arrow::array::{ Array, ArrayRef, ArrowPrimitiveType, A