News for hackers. Stuff that matters.
Re:Removing fsync from our local storage engine by zzsheng on Saturday May 09, 2026 @07:46PM (#48079477)
nop. we will not ack any write which is not in data or journal. please check the put details in the blog.
Re:Removing fsync from our local storage engine by loeg on Saturday May 09, 2026 @08:30PM (#48079759)
You initiate a write to the journal, but do not sync it before ACKing to the client.
Re:Removing fsync from our local storage engine by zzsheng on Saturday May 09, 2026 @08:36PM (#48079794)
journal file was pre-alloacated and we use direct-io for journal write so no need to call fsync.
Re:Removing fsync from our local storage engine by loeg on Saturday May 09, 2026 @09:27PM (#48080078)
Again, it is not durably persisted before acking to the client. Like I said earlier, that might be fine for your durability model, but it is unusual.
Re:Removing fsync from our local storage engine by thomas_fa on Saturday May 09, 2026 @11:52PM (#48080846)
We would wait for Bss data and journal DirectIO and the acking (sending response back to api_server) in the callback function. What you are implying is what s3 actually doing and you can get see from their paper[1] and we are stronger than that.

[1]https://www.amazon.science/publications/using-lightweight-fo....