srowen commented on issue #23411: [SPARK-26503][CORE] Get rid of 
spark.sql.legacy.timeParser.enabled
URL: https://github.com/apache/spark/pull/23411#issuecomment-452146452
 
 
   @MaxGekk I instead tried benchmarking old/new format, and 
`yyyy-MM-dd'T'HH:mm:ss.SSS` vs `yyyy-MM-dd'T'HH:mm:ss.SSSXXX` with input like 
`2011-12-03T10:15:30.000` and `2011-12-03T10:15:30.000+01:00`:
   
   Java 8:
   
   ```
   DateParsingBenchmark.newFormat_noZone    avgt    5  0.538 ± 0.066  ms/op
   DateParsingBenchmark.newFormat_withZone  avgt    5  0.693 ± 0.024  ms/op
   DateParsingBenchmark.oldFormat_noZone    avgt    5  0.745 ± 0.024  ms/op
   DateParsingBenchmark.oldFormat_withZone  avgt    5  0.760 ± 0.020  ms/op
   ```
   
   Java 11:
   
   ```
   DateParsingBenchmark.newFormat_noZone    avgt    5  0.579 ± 0.006  ms/op
   DateParsingBenchmark.newFormat_withZone  avgt    5  0.690 ± 0.026  ms/op
   DateParsingBenchmark.oldFormat_noZone    avgt    5  0.876 ± 0.009  ms/op
   DateParsingBenchmark.oldFormat_withZone  avgt    5  0.893 ± 0.036  ms/op
   ```
   
   No real difference (times are smaller here because I ran a shorter test.)
   
   Is that what you're interested in testing?
   
   If so are we back to favoring removing the old legacy parser? I'm OK with 
that.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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