LuciferYang commented on PR #37185: URL: https://github.com/apache/spark/pull/37185#issuecomment-1186257184
@srowen Summarized as follows： - Risk：The addition or deletion operation of `ArraySeq(both mutable and immutable)` will return a new `ArraySeq` and the underlying array cannot be grown in place. But there is also a risk: `mutable.ArraySeq.sortInPlace` will sort on the original array in place, which is a new method in Scala 2.13 and Spark doesn't use it. - Performance: The throughput of `arrayOfAnyAsSeq` scenario is `41.4M/s` before this pr, and its throughput is `2491.6M/s` after this pr. The performance data using Scala 2.12 are: `395.1M/s(Java 8)`, `1845.1M/s(Java 11)` and `1245.5M/s (Java 17)` - Whether `case array: Array[_] => array.toSeq.toArray[Any]` works： This is a TODO of mine, let me track it and if this is an issue, I will file a new Jira and try to solve it. -- 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]
