You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2021/11/26 09:33:35 UTC

[GitHub] [spark] zhengruifeng removed a comment on pull request #34367: [SPARK-37099][SQL] Impl a rank-based filter to optimize top-k computation

zhengruifeng removed a comment on pull request #34367:
URL: https://github.com/apache/spark/pull/34367#issuecomment-979657176


   test benchmark results on my laptop:
   
   ```
   
   [info] Java HotSpot(TM) 64-Bit Server VM 1.8.0_301-b09 on Linux 5.11.0-40-generic
   [info] Intel(R) Core(TM) i7-8850H CPU @ 2.60GHz
   [info] Benchmark Top-K:                          Best Time(ms)   Avg Time(ms)   Stdev(ms)    Rate(M/s)   Per Row(ns)   Relative
   [info] ------------------------------------------------------------------------------------------------------------------------
   [info] ROW_NUMBER WITHOUT PARTITION                       2368           2584         163          8.9         112.9       1.0X
   [info] ROW_NUMBER WITHOUT PARTITION (RANKLIMIT)           1019           1047          21         20.6          48.6       2.3X
   [info] RANK WITHOUT PARTITION                             2934           3087         142          7.1         139.9       0.8X
   [info] RANK WITHOUT PARTITION (RANKLIMIT)                   26             36          13        800.7           1.2      90.4X
   [info] DENSE_RANK WITHOUT PARTITION                       2762           2873         106          7.6         131.7       0.9X
   [info] DENSE_RANK WITHOUT PARTITION (RANKLIMIT)             22             28           4        934.7           1.1     105.5X
   [info] ROW_NUMBER WITH PARTITION                         19493          19758         319          1.1         929.5       0.1X
   [info] ROW_NUMBER WITH PARTITION (RANKLIMIT)              6373           6576         228          3.3         303.9       0.4X
   [info] RANK WITH PARTITION                               20696          21254         274          1.0         986.8       0.1X
   [info] RANK WITH PARTITION (RANKLIMIT)                    6172           6472         213          3.4         294.3       0.4X
   [info] DENSE_RANK WITH PARTITION                         19070          19346         284          1.1         909.3       0.1X
   [info] DENSE_RANK WITH PARTITION (RANKLIMIT)              6249           6439         131          3.4         298.0       0.4X
   
   
   
   ```


-- 
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: reviews-unsubscribe@spark.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org