2009/3/19 Gordan Bobic <gordan@xxxxxxxxxx>: > That's unavoidable to some extent, since the first server is the one that > is authoritative for locking. That means that all reads have to make a hit > on the 1st server, even if the data then gets retrieved from another server > in the cluster. Whether that explains all of the disparity you are seing, I > don't know. Not really. There is no need for locking while doing reads. Currently replicate will use the read-subvolume if specified, and the first available node if the option is not specified. Read load-balancing is in the pipeline, but it needs some other code changes to be implemented first. Vikas -- Engineer - Z Research http://gluster.com/
