We’ve an SSD pool and the initial Pool is created with 4K block sizes and the ZRAID is 10 with 12 disks.
Our VSMs is created with 4 K and the volumes is created a with record size of 8KB.
As we can see in the ElasticStore Graph the “Average Bblock size” is 8K.
The workload on the SSD Pool is 90% to run virtual machine on a XenServer with a Windows Server 2012 R2 OS.
Do you have any recommendations for us to saving space and still have enough performance, in terms of the pool/vsm block size and volume record size?
We already read the CloudByte Article about 4K Block sizes but at the and there is a recommendation to use 32K block sizes for the Pool – is this right?
You must be logged in to reply to this topic.