Performance Monitor and Remote Copies / Re-Striping

[Jim Silvia]

I just kicked off several large remote copies from one cluster to another cluster in the same management group.

I was monitoring performance and noticed very little throughput when monitoring the "Throughput Total" for the two node cluster.

I'm running ALB.

When i add each explicit storage node and view their "Throughput Total" the values are more in line with what i would expect.

See attached snapshot or if you just want the numbers

Cluster "Throughput Total" 7,131 B/s
NSM4 "Throughput Total" 28,320,255 B/s
NSM5 "Throughput Total" 28,412,224 B/s

Why doesn't the sum of the storage nodes equal the cluster?

[Jim Silvia]

I also discussed this with HP today.

They didn't give me a good answer. They were trying to say that this was the extra overhead that has to happen with Network Raid.

I think it's more of how they calculate/track - and that they exclude remote copies/striping in the cluster throughput total. It only seems to show ISCSI target related traffic.

As a test i simply launched a VM that was stored in a volume on this cluster and executed several large file copies. As soon as i did this the cluster throughput jumped as expected to come close to the individual nodes.