Created
July 19, 2010 15:01
-
-
Save PharkMillups/481511 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ronr_: well, was wondering if it's possible to configure riak to utilize different physical hard drives. | |
pharkmillups: ronr_ what do you mean exactly? | |
ronr_: pharkmillups: I think it may have to do with vnodes. since usually the | |
bottleneck is i/o, if I have multiple hard drives on the machine, it's possible to | |
somewhat release the bottleneck by allowing the app to write parts of the data to different disks. | |
drev1: ronr_: I think the only way for riak to take advantage of multiple disks on | |
a single machine would be to use "multi_backend" with each backend using a different | |
drive. You could then define on a per bucket basis which backend to use. | |
The problem with this setup, however, is that bitcask does not currently | |
support multi_backend functionality (https://issues.basho.com/210). | |
drev1: you could also run multiple nodes on a single machine but that | |
reduces fault tolerance | |
ronr_: drev1: thanks. | |
ronr_: which backend should give the best throughput? | |
justinsheehy: ronr_: under most circumstances, bitcask is easily | |
the highest-throughput backend | |
ronr_: justinsheehy: thanks... it took us a while to realize that it was the | |
REST-ful api that's making things terribly slow. | |
justinsheehy: ah. yeah, if throughput is a major goal, you want bitcask and you | |
want to use protobufs |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment