cigarl commented on issue #3954: URL: https://github.com/apache/iotdb/issues/3954#issuecomment-918777422
> But the key sentence is this. Even if the bug is fixed. Should we also call async every time when we flush log? Agree with you, in the IOT world, data may not be as sensitive,maybe we can sacrifice security _(in some cases)_ to improve performance _(we can even provide configurations to meet different needs of users)_.But for now, this approach seems to pose a greater danger _(eg. cluster can not be restarted successfully, or enter an unstable state, returning confusing results to the user, etc)_ in corner cases. BTW,maybe we should comb through some optimizations to compare.I'm so sorry that I seem to see only that the strategy of flushing to disk has an effect, the impact at the Raft level doesn't seem to be as large as expected. -- 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]
