option to use different disks for different vbuckets?
Mon, 11/22/2010 - 03:37
Consider the use-case with different datasets modeled as different vbuckets. The datasets are larger than RAM so fast disk-access is important. Each server (Amazon Ec2 m.xlarge) is configured with a couple of EBS-disks.
To me, it would make sense to be able to configure that each vbucket makes use of a different EBS-disk (or a cluster of RAID EBS's.) However I can't find the option to do so.
Is this somehow possible?
TIA,
Geert-Jan
Geert-Jan,
There is a way to map different datasets to disks. However, we have not tested this. So, pleae do the needed testing and let us know. Here is how you would do this:
You can specify buckets and create symlinks from the sqlite (C:\Program Files\Membase\Server\data\ns_1\yourbucket-*.sqlite ) files to your specific disks. Also, you would need to make sure that your sqlite journal mode is set to 'TRUNCATE' and not 'DELETE'.
PRAGMA journal_mode = TRUNCATE
Read more aboit Pragmas http://www.sqlite.org/pragma.html
Bhawana
Forum support is great for free but sometimes you need a guaranteed response time and dedicated resources for your questions or issues.
Consider purchasing enterprise-level support from Membase: http://www.membase.com/products-and-services/overview
Call or email "sales -at- membase -dot- com" today!