Speed up wal write
Registered by
Kostja Osipov
Right now we do 4 memcpy and 4 syscalls to make sure each user request which changes data reaches the disk.
Reduce this overhead by using shared memory and producer-consumer pattern, rather than the message-passing paradigm
used today.
Maintain "the last written lsn" across the front-end and the back-end, and alert all fibers waiting on wal-writer
based on the lsn they are waiting for (i.e. do not maintain an individual inbox in each fiber for that purpose).
Use offset_t as lsn, not consequetive lsn id -- that will simplify lsn generation.
Blueprint information
- Status:
- Complete
- Approver:
- Kostja Osipov
- Priority:
- Essential
- Drafter:
- Konstantin Shulgin
- Direction:
- Approved
- Assignee:
- Konstantin Shulgin
- Definition:
- Approved
- Series goal:
- Accepted for 1.4
- Implementation:
- Implemented
- Milestone target:
- 1.4.5
- Started by
- Kostja Osipov
- Completed by
- Kostja Osipov
Related branches
Related bugs
Sprints
Whiteboard
(?)