Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

If a shard goes down, writes to it are queued on a Gizzard middleware host. What happens if that Gizzard host is then lost?


The queues are local to each host so you'd lose those writes, but since writes are written to a journal on disk, you can apply some redundancy strategy to that as necessary. Depends on the SLA really.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: