peter-toth commented on code in PR #57346: URL: https://github.com/apache/spark/pull/57346#discussion_r3860919293
########## sql/core/benchmarks/WindowBenchmark-results.txt: ########## @@ -2,171 +2,287 @@ Section A - MIN (non-invertible) ================================================================================================ -OpenJDK 64-Bit Server VM 17.0.20+8-LTS on Linux 6.17.0-1020-azure +OpenJDK 64-Bit Server VM 17.0.20+8-LTS on Linux 6.17.0-1022-azure AMD EPYC 7763 64-Core Processor MIN sliding window, W=1001, 256K rows: Best Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative ------------------------------------------------------------------------------------------------------------------------ -MIN naive (current, baseline) 4014 4035 17 0.1 15313.6 1.0X -MIN segtree (default) 392 407 16 0.7 1496.8 10.2X -MIN segtree (blockSize=256) 2199 2214 12 0.1 8388.4 1.8X +MIN naive (current, baseline) 4036 4059 17 0.1 15396.6 1.0X +MIN segtree (default) 377 390 13 0.7 1439.8 10.7X +MIN segtree (blockSize=256) 2220 2226 6 0.1 8470.2 1.8X +MIN monotonic deque (new) 120 122 3 2.2 458.3 33.6X Review Comment: Correction on the provenance, which makes this a bit sharper than I put it: `bf07c13` is not a fresh regeneration, it is the rebase of the R4 results commit `7bf57e5` - the file is byte-identical between the two. So this artifact is the one generated back at R4, and it is now two code commits behind (`e02bbe4`, `b7293b4`). The measurement and the ask are unchanged. -- 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]
