What kind of workload do you have? Why are you using stripe? That is a 14 nodes cluster doing a lot of analyse in R or with classical tools in the biology. these computing used very large files from 30Go to 1To, it is manage with a sheduler. and interconnected with a ten Gbs. So I think that distributing data on differents nodes will "equalizes the access" for all nodes. But maybe it's a bad idea What are your feeling on that massive data access in a gluster volume ? Pierre Léonard |
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://www.gluster.org/mailman/listinfo/gluster-users