I'm really excited about this next option in our mdisk group analysis section. It is the first time that we can analyse into the internal SVC queues down to the storage system.
With this new option we get a much better view of the performance problems of the storage system.
It is obvious that this backend storage system was in deep performance troubles from 5 am to 6:30 am.
Have a look to this:
The picture belongs to this whitepaper
How destage rates react on cache partition fullness
Pict4: MDISK GROUP: We have a deeper look into the backend storage with the latest BVQ Version. Since BVQ 2.9 we split the mdisk Group latency into above queue latency, queue latency and below queue latency. This is a very powerful feature because here we can see that the backend storage system was not at all able to handle the load from the SVC in the time period from 6am to 6:30am. The above queue latency is much higher than the below queue latency - in other words - data is sitting and waiting in the SVC queue to be transported to the backend storage. This shows that the backend system is not able to handle the load. From 6:30am on the above and below queue latency is the same - so the SVC queue was near to be empty because the storage system could handle the load.