Originally posted by: sanation99
Hi hoping someone can enlighten me :-) we've recently put in a new HDS SAN, a VSP with an AMS2500 behind it providing the storage (SAS), configured into DP polls on the VSP. We're all ESX/VM at the front. One of the environments we've put in is a VDI desktop system. This has been in a week with no issues until a couple of days ago when we started getting high latencies on our writes with peaks around 160 ms! The only difference being we've just switched on replication to a same system about 60km away. But we're using HUR which I thought wouldn't have effected it, but it does seem to point to the replication - we had the issues once we turned it on, and now we've stopped replication ( LUNs gone back to simplex) the issues have rectified. Checking the alerts we only had one instance of the jpurnal volume filling up. Should HUR have any effect on the source LUN performance?