srowen commented on issue #25279: [SPARK-28519][SQL] Use StrictMath log, pow 
functions for platform independence
URL: https://github.com/apache/spark/pull/25279#issuecomment-519195159
 
 
   @gatorsmile here's a quick benchmark for `log()`:
   
   Previous:
   ```
   Running benchmark: log
     Running case: log
     Stopped after 20 iterations, 6376 ms
   
   Java HotSpot(TM) 64-Bit Server VM 1.8.0_192-b12 on Mac OS X 10.14.6
   Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz
   log:                                      Best Time(ms)   Avg Time(ms)   
Stdev(ms)    Rate(M/s)   Per Row(ns)   Relative
   
------------------------------------------------------------------------------------------------------------------------
   log                                                 297            319       
   17          6.7         148.3       1.0X
   ```
   Current:
   ```
   Running benchmark: log
     Running case: log
     Stopped after 20 iterations, 6507 ms
   
   Java HotSpot(TM) 64-Bit Server VM 1.8.0_192-b12 on Mac OS X 10.14.6
   Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz
   log:                                      Best Time(ms)   Avg Time(ms)   
Stdev(ms)    Rate(M/s)   Per Row(ns)   Relative
   
------------------------------------------------------------------------------------------------------------------------
   log                                                 301            325       
   15          6.7         150.3       1.0X
   ```
   
   Looks like it's slower by about 2 nanoseconds, but it's not even 
statistically significant.
   Even StrictMath is native code. I think we could say there is no perf hit.

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to