HPE MSA 2052 "virtual storage?"

[avicci]

Need some clarify from what I'm reading in the documentation. It seems data is stored across all disk groups in a virtual pool. Meaning, If I built one RAID 1 disk group, and another RAID 5 disk group with the intention of splitting SQL transaction logs from the database, this would give me no benefit as I don't have control where the files are actually stored on the hardware?

[support_s]

System recommended content:

1. Advisory: HPE MSA 1040, 2040, 2042, 1050, 2050, and 2052 SAN Storage - Migration Of Virtual Disk Groups From MSA To Another HPE MSA With Configured Virtual Disk Groups Is Not Supported

2. HPE MSA 2052 SAN Storage

Please click on "Thumbs Up/Kudo" icon to give a "Kudo".

Thank you for being a HPE valuable community member.

[sbhat09]

Hello @avicci,

That is not an ideal way. RAID-6 is best recommended for SQL or any database volumes.

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of the company.

[JonPaul]

@avicci 
You have correctly interpreted how Virtual Storage functions.
When you create disk-groups in a Pool that will be in the same Tier (Archive, Standard or Performance) they should be as close to identical as possible to be of equal performance.
The automated tiering engine will place pages of data where it make sense from a Tier perspective, highly accessed pages will migrate to the fastest available tier, but does not track performance capabilities of disk-groups within the tier.  So in a case where you have a RAID 1 and RAID 5/6 in the same tier, you will get one page of data from the 2 drives of RAID 1 (slow) and the next page of data from the many drive RAID 5 (fast).
When you have an application which 'requires' different storage capabilities you can use the 2 Pools and create one Pool for one workload type and the other Pool for the other workload type.  If that is not an option you can also utilize the Tier Affinity on a volume option which can give indications of what tier the data should land on and the RAID level of that Tier.  Ex for your dataset/random workload set the affinity to performance (performance tier not required) which will set the data to be 'preferred' into the fastest available tier.  Set the log/sequential workload to Archive with an Archive Tier and RAID 10 disk-group.  The above example would 'prefer' the data to the correct tier and RAID level but is not guaranteed.
The last option, and easiest, is to utilize RAID 6 for all the workloads.  The cache available in the MSA controller tends to lessen the 'requirement' of differing RAID levels for different workloads.
RAID 6 for the redundancy.
Hope this helps.