We're currently running a pair of FAS8020s, running OnTap 8.3.2P6 as NFS shares with quite a large number of files per volume. These are then accessed by Linux clients via NFS4. The primary connection for these is via 1GB copper, due to the existing infrastructure. At certain times of day we're seeing performance drop on one or more of the shares when the number of concurrent access requests to different parts of the storage increase.
A number of different solutions have been attempted internally, and when we sideways scaled the number of clients some of the big hits to performance were negated. But, as with anything, there’s only a certain amount of sideways scaling that’s possible.
Has anyone experienced similar issues previously, either with these specific devices or with this configuration?
better to involve ps team/open a case to analyse the performance bottleneck. then to figure out the best way
Cannot find the answer you need? No need to open a support case - just CHAT and we’ll handle it for you.