Re: Volume Tiering

[Lindsay Mathieson <lindsay.mathieson@xxxxxxxxx>]

On 2/05/2016 10:15 PM, Mohammed Rafi K C wrote:

- I presume the files are promoted across all bricks. i.e you can't have different files promoted per brick.

I didn't get your question correctly. But I will try to answer generically . File movement happens from one tier to another tier, as I mentioned earlier, you can consider this as moving one file from one volume to another volume.

I expressed it somewhat poorly. I was simplistically thinking of a one to one mapping of a Replica 3 cold tier to a replica 3 hot tier on a 3 node compute cluster (basically what we have). Access to files would only happen locally on the node, so it would be more efficient to just promote the files that were accessed locally, which would vary from node to node.

But that would defeat the whole concept of a  gluster tier :) what I am describing is really just a simple SSD cache, which I already have with the underlying ZFS pool with a SSD cache. So please ignore the question, it was poorly thought out in the first place.

As an example, let us assume, you are moving a file from hot tier (replica configuration) to a cold tier which is a distributed disperse configuration, then the file will be hashed to any of the distribute set, then the file will be created with ec meta data (data and parity).

I'm  not sure, I made it more complicated ;).

No, it make perfect sense thanks.

TBH, I'm not sure that with my setup - Bricks on ZFS pool with SSD logs & cache that a SSD tier would be a big improvement. It shall be interesting to compare.

Cheers,

-- 
Lindsay Mathieson

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://www.gluster.org/mailman/listinfo/gluster-users