MSA 2052 Storage allocation query

[MikeH71]

Hi, I'm looking for some advice on best practice on an MSA 2052 and expander.

The MSA has 2 x 800GB SSD and 22 x 1.8TB SAS drives.

The expansion cage has 12 x 4TB MDL drives.

The SAN is SAS-connected to an HPE ML350 Gen9 using HBA and all of this storage is to be used for Windows files.

The 2052 is RAID 10 and will present one "drive" to a Windows server.

The expansion is RAID 6 and will present as another drive in Windows server.

My question is regarding maximum volume and pool recommendations to allow one snapshot on each pool. What would be the recommended setup for this? Before the expansion cage was installed today, we had one pool covering the performance cache and 10 SAS drives (volume A, 8TB), and another covering the second set of 12 SAS drives (volume B, 9TB), but this seems too high I think.

Many thanks

[SUBHAJIT KHANBARMAN_1]

If you want to know about max limits then I would suggest to go through Quick Spec as below, https://h20195.www2.hpe.com/V2/getpdf.aspx/a00008277enw.pdf

Snapshots reference volumes, they take up no space unless they or the source volume or source snapshot is modified. Space does not have to be reserved for snapshots because all space in the pool is available for them.

So you can create snapshots for each volumes you have created in a specific Pool. It means Pool A Volumes Snapshots and Pool B Volume Snapshots both will take space from respective pools only.

If you want to take the Advantage of Three Tier Architecture like Performance, Standard and Archive tier then put all drives under single Pool. Otherwise you can put SSD and SAS drives in one Pool and all MDL drives in other Pool. You can also put SSD, few SAS drives and few MDL drives in one Pool. Rest of the SAS drives and MDL drives in other Pool. This will give one Pool wth Three Tier and other Pool with twp Tier advantage. It depends upon your business requirement and application that you want to run on which type of drives.

Hope this helps!
Regards
Subhajit

I am an HPE employee

If you feel this was helpful please click the KUDOS! thumb below!

*************************************************************************

[MikeH71]

Thanks for the reply.

The use of this is for archive data (MDL drives) and live project data (SSD and SAS).

The live project data is changing frequently and this could be the additon and/or removal of between 5 and 8TB of data 2 or 3 times per week. This has caused issues with the snapshots before so I want to make sure we set it up right so we don't have issues in the future.

The live data is expected to be around 10TB at any time, but changes multiple times per week. Our backups are affected because of this frequency of change. 

If you have a specific recommendations I'd be keen to know.

Thanks again

[SUBHAJIT KHANBARMAN_1]

I would suggest you to go with three tier or Automated Tier setup where you will have SSD, SAS and MDL drives. 

In your case let's say you have 10TB Thin Volume and out of that around 6TB data is there. You have taken snapshot of this. Then you may have added 1TB data so now you have 7TB data. You have taken another snapshot now. Then you have removed 3TB data. Then you have taken another snapshot.

The good part of this MSA, unlike previous model or any other array - MSA2052 snapshot or Virtual Snapshot technology works in Redirect on Copy (ROW) rather than Copy on Write (COW) technology. Here all will be part of same Pool so there is no need to copy any data for snapshot. All will be poining to the data in same pool. So you shouldn't face any challange.

Hope this helps!
Regards
Subhajit

I am an HPE employee

If you feel this was helpful please click the KUDOS! thumb below!

************************************************************************