stevomitric commented on PR #56744: URL: https://github.com/apache/spark/pull/56744#issuecomment-4799376081
> Observation 1: numIters=1 for write benchmarks Intentional - this mirrors the IO-benchmark analogue DateTimeRebaseBenchmark, which uses 1 iteration for Save cases and 3 for Load. > Observation 2: Coverage gaps for a comprehensive nanos-vs-micros comparison joins/filter/window do work today because comparator/hashing, but they are not tested so i would leave them for a later followup. > Observation 3: PR description says "p in [7, 9]" but only p=9 is benchmarked. added p=7 cases for both parse and cast-to-string -- 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]
