zhuqi-lucas opened a new pull request, #13942: URL: https://github.com/apache/datafusion/pull/13942
## Which issue does this PR close? Closes [#13941 ](https://github.com/apache/datafusion/issues/13941) ## Rationale for this change ## What changes are included in this PR? ## Are these changes tested? yes: ```rust cargo run --release --bin dfbench clickbench --iterations 5 --query 35 --explain Finished `release` profile [optimized] target(s) in 0.18s Running `target/release/dfbench clickbench --iterations 5 --query 35 --explain` Running benchmarks with the following options: RunOpt { query: Some(35), common: CommonOpt { iterations: 5, partitions: None, batch_size: 8192, debug: false }, path: "benchmarks/data/hits.parquet", queries_path: "benchmarks/queries/clickbench/queries.sql", output_path: None, explain: true } Q35: SELECT "ClientIP", "ClientIP" - 1, "ClientIP" - 2, "ClientIP" - 3, COUNT(*) AS c FROM hits GROUP BY "ClientIP", "ClientIP" - 1, "ClientIP" - 2, "ClientIP" - 3 ORDER BY c DESC LIMIT 10; Query 35 iteration 0 took 1202.7 ms and returned 10 rows Query 35 iteration 1 took 1092.0 ms and returned 10 rows Query 35 iteration 2 took 1021.7 ms and returned 10 rows Query 35 iteration 3 took 1022.2 ms and returned 10 rows Query 35 iteration 4 took 982.8 ms and returned 10 rows +---------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ | plan_type | plan | +---------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ | logical_plan | Sort: c DESC NULLS FIRST, fetch=10 | | | Projection: hits.ClientIP, hits.ClientIP - Int64(1), hits.ClientIP - Int64(2), hits.ClientIP - Int64(3), count(*) AS c | | | Aggregate: groupBy=[[hits.ClientIP, __common_expr_1 AS hits.ClientIP - Int64(1), __common_expr_1 AS hits.ClientIP - Int64(2), __common_expr_1 AS hits.ClientIP - Int64(3)]], aggr=[[count(Int64(1)) AS count(*)]] | | | Projection: CAST(hits.ClientIP AS Int64) AS __common_expr_1, hits.ClientIP | | | TableScan: hits projection=[ClientIP] | | physical_plan | SortPreservingMergeExec: [c@4 DESC], fetch=10 | | | SortExec: TopK(fetch=10), expr=[c@4 DESC], preserve_partitioning=[true] | | | ProjectionExec: expr=[ClientIP@0 as ClientIP, hits.ClientIP - Int64(1)@1 as hits.ClientIP - Int64(1), hits.ClientIP - Int64(2)@2 as hits.ClientIP - Int64(2), hits.ClientIP - Int64(3)@3 as hits.ClientIP - Int64(3), count(*)@4 as c] | | | AggregateExec: mode=FinalPartitioned, gby=[ClientIP@0 as ClientIP, hits.ClientIP - Int64(1)@1 as hits.ClientIP - Int64(1), hits.ClientIP - Int64(2)@2 as hits.ClientIP - Int64(2), hits.ClientIP - Int64(3)@3 as hits.ClientIP - Int64(3)], aggr=[count(*)] | | | CoalesceBatchesExec: target_batch_size=8192 | | | RepartitionExec: partitioning=Hash([ClientIP@0, hits.ClientIP - Int64(1)@1, hits.ClientIP - Int64(2)@2, hits.ClientIP - Int64(3)@3], 14), input_partitions=14 | | | AggregateExec: mode=Partial, gby=[ClientIP@1 as ClientIP, __common_expr_1@0 - 1 as hits.ClientIP - Int64(1), __common_expr_1@0 - 2 as hits.ClientIP - Int64(2), __common_expr_1@0 - 3 as hits.ClientIP - Int64(3)], aggr=[count(*)] | | | ProjectionExec: expr=[CAST(ClientIP@0 AS Int64) as __common_expr_1, ClientIP@0 as ClientIP] | | | ParquetExec: file_groups={14 groups: [[Users/zhuqi/arrow-datafusion/benchmarks/data/hits.parquet:0..1055712604], [Users/zhuqi/arrow-datafusion/benchmarks/data/hits.parquet:1055712604..2111425208], [Users/zhuqi/arrow-datafusion/benchmarks/data/hits.parquet:2111425208..3167137812], [Users/zhuqi/arrow-datafusion/benchmarks/data/hits.parquet:3167137812..4222850416], [Users/zhuqi/arrow-datafusion/benchmarks/data/hits.parquet:4222850416..5278563020], ...]}, projection=[ClientIP] | | | | +---------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ ``` ## Are there any user-facing changes? yes -- 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: github-unsubscr...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
