Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

rocksdb: 1-write n-read, engine level distribution #8

Open
4 tasks
rockeet opened this issue Jun 7, 2018 · 1 comment
Open
4 tasks

rocksdb: 1-write n-read, engine level distribution #8

rockeet opened this issue Jun 7, 2018 · 1 comment
Assignees
@mm304321141

Comments

@rockeet
Copy link
Member

rockeet commented Jun 7, 2018 •
edited
Loading

  • sync in-memory change
  • load sst after compaction/flush
  • since writer/master may delete file if not needed any more, so it must "know" files in using by reader/slave
    • explicit communication between writer & reader is required
    • tailing on wal-log is simpler & slower than ZeroMQ
    • manifest files also need to be monitored
  • there are two kind of readonly instance
    • readonly & sync writer's update on DB, used for OLTP
    • readonly & do not sync writer's update, used for OLAP(spark...), and distributed compaction
@rockeet
Copy link
Member Author

rockeet commented Jun 8, 2018

zstd compressed dictionary load maybe a bottle neck. how to tradeoff?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@mm304321141 mm304321141

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@mm304321141 @rockeet