• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 787
  • Last Modified:

SAN Configuration

I have two servers with two single HBA's each.
Two Fibre Switches (Not linked) one an 8Gb the other a 4 Gb. Both with 4 spare ports.
One IBM 3524 storage array with dual controllers each controller with two ports.

I am trying to come up with a Zoning config that will accomodate my implementation plan which is:

4 Drives (Raid 10) Shared clustered volume to be accessed by both servers for hyper failover clustering. (All one lun)

8 Drives (Raid 5) split into two luns. (File shares only) One hot spare to cover both arrays.

My objectives are to have as much redundacy and throughput as possible on all connection and would like some advice on fibre links, Zoning, and whether roundrobin or shortest queue is preferable for what I am setting up?
2 Solutions
SHBStorage Network SpecialistCommented:
Is your HBA in each servers with dual ports? If not you'd not be able to achieve redundancy

Add dual port HBAs in the servers and connect each ports to both fabrics if they are in redundant. And connect from each storage controller to both the fabrics.

Zone the servers to storage controllers in each fabrics. In the server add MPIO feature. Round robin would be ideal parameter.

Hope this clarifies
They have two single port HBAs so they can achieve redundancy, a little bit superior to a dual port HBA since it covers HBA failure as well as path failure.

Best practice is to use single initiator zoning, so you would have a zone for each HBA that includes the HBA's port plus the port on each controller. Look at http://www.redbooks.ibm.com/redbooks/pdfs/sg247914.pdf fig 3-9 one zone would be host 1 HBA1 plus the light green and pink conected ports on the SAN storage. (ignore the fact that they have drawn dual port HBAs and pretend it shows two single port HBAs instead - cabling and zoning is the same. Zone2 would be server 1 HBA2 (grey cable) plus dark green and light blue SAN storage ports, repeat for server2. Total 4 zones.

For MPIO you will have 4 paths, the manual above says to use round robin. Since each LUN is owned by a particular controller until failover occurs you should spread LUN ownership between the two controllers so they have half of the LUNs each, in your case I would guess that Hyper-V VMs will have more load than file serving so hyper-v cluster owned by one controller and both file server LUNs owned by the other.

Note you don't need switches for a two host redundant configuration, you could connect HBAs direct to each controller if the HBAs support loop topology (Brocade ones don't yet AFAIK). Still, you already have the switches and you may want to add more hosts later.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

Join & Write a Comment

Featured Post

Free Tool: ZipGrep

ZipGrep is a utility that can list and search zip (.war, .ear, .jar, etc) archives for text patterns, without the need to extract the archive's contents.

One of a set of tools we're offering as a way to say thank you for being a part of the community.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now