hangc0276 commented on issue #2395: URL: https://github.com/apache/bookkeeper/issues/2395#issuecomment-982273158
After deep into the compaction code, i found there is no need to divide `compactionRateByBytes` into `compactionWriteRateByBytes` and `compactionReadRateByBytes` https://github.com/apache/bookkeeper/blob/abf1a96e38bb74f7f7a47e9b6f2b337bd07d622f/bookkeeper-server/src/main/java/org/apache/bookkeeper/bookie/EntryLogger.java#L1009-L1051 The reason is 1. The entry log scanner will just read each entry's metadata and decide whether to replicate it. If no need to replicate, it will skip the entry data. So there will be no heavy read data effects. 2. The read entry from entry log and write to new entry log logic is in a while loop. Currently, we throttle on write entry side, it will also throttle the entry read. -- 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]
