zzsheng

we manage memory instead of relying on page cache

sorry, typo. data will *not* be lost.

check s3 express one zone

journal file was pre-alloacated and we use direct-io for journal write so no need to call fsync.

nop. we will not ack any write which is not in data or journal. please check the put details in the blog.

actually both crash consistentcy and durability. after we ack, we make sure data will be lost due to crash, restart or power loss.

we also use direct-io for reads.

thanks for feedback. actually it was pointed out in blog that we do not use append only log to avoid fsync due to size change. what we use is preallocate fixed size log file and we do write journal data and space reclaim by 4KB unit, also with direct-io.

Author here. This is not a general argument against fsync; the design depends on SSD-only deployment, preallocated files, O_DIRECT, single-key atomicity, and device write guarantees.