first of all, I’m on a dev-server so disks arent optimized (7200RPM consumer rate disk)
Still when inserting at a rate of ~2000 docs/sec (about 10 fields, say 2KB per doc) I see the queue depth steaditly climbing. Logically since the drainrate is ~100 docs/sec on average. (With distribution all over the place from range ~ [30, 250] docs/sec.
Still strikes me as incredibly low correct? Any ballpark figures of what I might be expecting?
Hello,
Which version of Couchbase are you using? (2.0.1 or 2.1)
Which OS?
How to you insert these documents?
Can you check that the Couchbase process (memcached) is writing on disk “all the time” ?
If you are using 2.0.1, can you switch to the 2.1 release?
In 2.1 it is possible to configure the number of thread for read and write, this will help us to see what is the limiting factor in your environment.
see http://www.couchbase.com/docs/couchbase-manual-2.1.0/couchbase-admin-tasks-mrw.html
Regards
Tug
@tgrall
Hello,
Have you tested with the new Couchbase 2.2 release?
And add more workers for you cluster:
http://docs.couchbase.com/couchbase-manual-2.2/#couchbase-admin-tasks-mrw
Regards
Tug
@tgrall