gortiz commented on PR #8979: URL: https://github.com/apache/pinot/pull/8979#issuecomment-1234453883
These are the results I'm having with the last version. It seems that the descending version is slower than in the first attempts. It may be due to the incorrect behaviors that were detected by @Jackie-Jiang, but it may also be some implementation inefficiencies. I'll will try to study that a bit ``` Benchmark (_numRows) (_primaryRepetitions) (_scenario) (_skipOptimization) Mode Cnt Score Error Units BenchmarkOrderByQueries.sortedAsc 1500000 1 EXP(0.5) true avgt 5 36.699 ± 1.108 ms/op BenchmarkOrderByQueries.sortedAsc 1500000 1 EXP(0.5) false avgt 5 1.349 ± 0.091 ms/op BenchmarkOrderByQueries.sortedAsc 1500000 1000 EXP(0.5) true avgt 5 29.210 ± 1.966 ms/op BenchmarkOrderByQueries.sortedAsc 1500000 1000 EXP(0.5) false avgt 5 0.421 ± 0.032 ms/op BenchmarkOrderByQueries.sortedAscPartially 1500000 1 EXP(0.5) true avgt 5 85.217 ± 11.162 ms/op BenchmarkOrderByQueries.sortedAscPartially 1500000 1 EXP(0.5) false avgt 5 1.460 ± 0.261 ms/op BenchmarkOrderByQueries.sortedAscPartially 1500000 1000 EXP(0.5) true avgt 5 81.292 ± 25.354 ms/op BenchmarkOrderByQueries.sortedAscPartially 1500000 1000 EXP(0.5) false avgt 5 2.086 ± 0.500 ms/op BenchmarkOrderByQueries.sortedDesc 1500000 1 EXP(0.5) true avgt 5 352.122 ± 37.947 ms/op BenchmarkOrderByQueries.sortedDesc 1500000 1 EXP(0.5) false avgt 5 14.572 ± 0.771 ms/op BenchmarkOrderByQueries.sortedDesc 1500000 1000 EXP(0.5) true avgt 5 283.754 ± 12.982 ms/op BenchmarkOrderByQueries.sortedDesc 1500000 1000 EXP(0.5) false avgt 5 9.702 ± 2.364 ms/op BenchmarkOrderByQueries.sortedDescPartially 1500000 1 EXP(0.5) true avgt 5 443.211 ± 53.459 ms/op BenchmarkOrderByQueries.sortedDescPartially 1500000 1 EXP(0.5) false avgt 5 71.102 ± 64.952 ms/op BenchmarkOrderByQueries.sortedDescPartially 1500000 1000 EXP(0.5) true avgt 5 623.616 ± 45.834 ms/op BenchmarkOrderByQueries.sortedDescPartially 1500000 1000 EXP(0.5) false avgt 5 92.627 ± 9.508 ms/op ``` -- 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: commits-unsubscr...@pinot.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org For additional commands, e-mail: commits-h...@pinot.apache.org