Bottleneck-our Hyper-V virtual environment

Kelvin.uk — Tue, 24 Jan 2012 04:42:23 GMT

Hi,

We have a bottleneck no our Hyper-V virtual environment, and I'm not sure what is causing the problem.

I was trying to migrate SRV1 from LUN2 to LUN3 using System Center Virtual Machine Manager 2008. At the time only DC1 and SCVMM VMs where switched on. It seems to transfer at a reasonable rate at first, transferring about 40GB in 30-45 minutes. Then once it reaches the 40GB mark the transfer rate completely drops. When the transfer rate dropped I cleared the port summary on both switches and monitored the transmitted and received pause frames for 20 minutes. I did not see any.

Some figures on just how poor the moving data from LUN2 to LUN3 was:

Really need to know what's causing the bottleneck here! Here our the specs, anything else you need to know just ask.

HP MSA2324i

LUN1: 6x3 7.2k RAID10

LUN2: 6x3 7.2k RAID10

LUN3: 6x3 10k RAID10

2 x HP Procuve 1800

Flow control enabled, jumbo packets enabled, switches interconnected with trunks etc...

2 x HP DL380G7

92GB, 24CPU count per host

Windows Server 2008R2

The environment was professional setup and I believe the bottleneck is due to hardware limitation, rather than incorrect setup. I had to cancel the storage migration in the end as there was simply not enough time to complete this.

Re: Bottleneck-our Hyper-V virtual environment

Kelvin.uk — Tue, 24 Jan 2012 14:46:04 GMT

No feedback yet, no ideas at all?

To sum this up I would like to know why it took 13 hours to migrate just 145GB from LUN2 to LUN3, I work that out as 0.18GB a minute! No other VMs where switched on either.

topic Re: Bottleneck-our Hyper-V virtual environment in MSA Storage

Bottleneck-our Hyper-V virtual environment

Re: Bottleneck-our Hyper-V virtual environment