Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

If so how are you finding it and how do you have your multi-site setup configured?

We're looking more and more likely to go the P4000 route and I'm looking at how things like failover work between the P4000 and vSphere (I have read the HP PDF's).

share|improve this question
I want a trio of those desperately, but want does not equate have, alas. – sysadmin1138 Oct 9 '10 at 16:07
Had a senior lefthand guy come to talk to us to back up the evaluation we're doing - came away very impressed just trying to tidy up some loose ends. – Hutch Oct 9 '10 at 16:23
up vote 2 down vote accepted

Our group has four P4000s across two sites. We use them to host HA VMs on Hyper-V. VM Guest performance has been very good. Using multipath IO on 1 Gb ethernet SAN fabric, with jumbo frames and flow control, per HP guidance. I'm guessing performance would be similar for vSphere HA. Failovers with live migration is a non-event.

To perform a multi-site setup, you need to have nodes at either end. We use async replication, because our link is not fast enough for synchro rep (needs to be 1 Gb or greater). Still suits us for DR purposes. Eventually, we'd like to implement synchro. The management console is a breeze to use.

We bought our initial applicances back when it was still LeftHand Networks. We've compared Dell EqualLogic and felt this served our needs better.

share|improve this answer
Thanks, still waiting for stuff to arrive but we went the P4000 route in the end, we felt it offered more flexibility than Equallogic and some other options (VSA's etc.). – Hutch Nov 9 '10 at 17:42

I would highly recommend you read this guys post.

We also has similar problems with HP support, and some problems with the P4500s crashing for no apperent reason. I guess thats why you need two of them all the time.

I inherited this LH stuff when i joined the company. We stopped using it about 6 months ago. You have to understand the reason that no other vendors use commodity x86 servers to build SANs with.

Its a fundamental architectural weakness. based on my years of experience SANs

share|improve this answer
That's not quite true - Compellent use more or less commodity x86 servers too if I read their specs correctly. But the point is accurate - each node is a more or less standard server and all resilience requires multiple P4000's, and they will need a third node (usually a virtual appliance) to act as a failover manager if you only have two arrays. – Helvick Dec 12 '10 at 15:49

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.