Hello all,
We've got an interesting problem of performance and was wondering if anyone else had encountered the same.
Simply put, we've got two physical servers forming a cluster. On the cluster we have several sap zones: one zone is a single zone (ie sap and ora running in the same zone)with failover capabilities (ie it can fail across to the other physical node), another zone is a clustered pair - either of the two zones - one running sap, one running ora - can fail over to the partner server. This is Solaris 10 , Suncluster 3.2.
The problem: the single zone sap instance writes data to the backup media server at 40m/sec. But the clustered pair can only manage 16m/sec (to the same media server): I would have imagined that the pair of zones would be able to perform at least as fast as the single zone, but this is not true:its markedly worse. The problem is basically the same across all services where we have a two zone cluster compared to a single floating zone.
I realise its not exactly a netbackup issue, however if its been seen as a netbackup issue, then theres a chance that a solution exists (most likely a Solaris / Cluster fix).
Thanks in advance,Tim
