wiki

For information, for volume mirroring, the max speed is sync rate 100% corresponding to 64MB/s

Change perf collect interval, interval of 5 minutes give a better view.

V5200 > startstats -interval 5

From the provided data the average write response time for vdisk is below 2,5 ms . Vdisk id 45 with highest average response times shows as well high write data rate and io rate for a specific time frame leading to a higher write response time on node level too.

I would suggest to implement a throttle for the volume to limit the IO a bit and see if this helps.

n regards to your question for data throughput limit the answer is:that compressed volumes works best with throughput below 350MB/s. The data below shows that for V5200-2 you see a bit higher response times as the throughout is about 400 MB/s. If you want to achieve a bit better response time you may want to throttle the throughput to less than 350MB. The command is described here.

https://www.ibm.com/docs/en/flashsystem-5x00/8.3.1?topic=csc-mkthrottle-3

Example :

V5200 > mkthrottle -type vdisk -bandwidth 300 -vdisk 49

Enable the host unmap feature, using either the chsystem command line. Note: For systems running 8.2.1 or higher

V5200 > chsystem -hostunmap on

The host unmap feature in Spectrum Virtualize allows host filesystems to inform the storage system that a region of storage is no longer required and can be cleaned up to make space for new data to be stored. It is highly desirable when using data reduction (FCMs or Data Reduction Pools) to avoid running out of space.

Best way is to addvdiskcopy, which is a parallel process and can be slow down this sync rate parameter, while movevdisk is much more slower whith 1 or 2 vdisk movement at same time

Same command as movevdisk:

SVC > addvdiskcopy -mdiskgrp mdisk_group_name vdisk_name -autodelete yes 

sg247521 - IBM System Storage SAN Volume Controller Best Practices and Performance Guidelines.pdf (www.redbooks.ibm.com) p61

When you plan to attach a V7000 on the SAN Volume Controller, create the arrays (MDisks) manually (by using a CLI), instead of using the V7000 settings. Select one disk drive per enclosure. When possible, ensure that each enclosure that is selected is part of the same chain. When you define V7000 internal storage, create a 1-to-1 relationship. That is, create one storage pool to one MDisk (array) to one volume. Then, map the volume to the SAN Volume Controller host.

Important:The extent size value for SAN Volume Controller should be 1 GB. The extent size value for the V7000 should be 256 MB. These settings stop potential negation of stripe on stripe. For more information, see the redbook (previously mentioned) and the blog post “Configuring IBM Storwize V7000 and SVC for Optimal Performance” at:
https://www.ibm.com/developerworks/mydeveloperworks/blogs/storagevirtualization/entry/configuring_ibm_storwize_v7000_and_svc_for_optimal_performance_part_121?lang=en

Best practice on host is to use 4 paths with a maximum of 8 paths per volume (LUN). Depending on read/write block, you will have a decrease of IOPS from 0 to 10% when you goes from 4 to 8 paths !

you need at least 4 arrays per V7000 to get maximal CPU core performance.

In other words, if you only run one array, you will get 1/4 of the performance of the system. Optimal configurations require at least 4 arrays, to ensure maximal core usage. Since arrays can be accessed through both nodes in the V7000 system, 4 is enough, on each node, one array will be assigned and processed through one core. http://rogerluethy.wordpress.com/2012/04/20/configuring-ibm-storwize-v7000-and-svc-for-optimal-performance-part-1/

The cluster has detected that an IBM DS series disk controller's configuration is not supported by the cluster. The disk controller is operating in RDAC mode. The disk controller might appear to be operating with the cluster; however, the configuration is unsupported because it is known to not work with the cluster. User response

  Using the IBM DS series console, ensure that the **host type** is set to **IBM TS SAN VC'** and that the AVT option is enabled. (The AVT and RDAC options are mutually exclusive).
  Mark the error that you have just repaired as "fixed". If the problem has not been fixed it will be logged again; this could take a few minutes.
  Go to repair verification MAP.