Throughput question / vmware datastore configuration

Posted on 2014-08-14
Last Modified: 2014-08-18
We have an ML350 G6 running VMware 5.x with a single datastore - two virtual machines, both Win2008 servers - one a DC/filer server and the other a low-use SQL 2008 server.

The server had 4x300GB SAS 10K drives configured in a RAID 5 on the internal SmartArray P410i/256MB controller.

An urgent, unexpected need came up for more storage space so we added 4 more 300GB 10K SAS drives, configured them for RAID5 again on the same controller (we couldn't afford the space for anything other than RAID5) and SPANNED the datastore across this new LUN.

It seems that, more times than not, throughput - especially during backups - is especially hampered.  Average latency is HIGH - approaching 45 ms sometimes.  We're experiencing problems the VMs becoming unresponsive when consolidating snapshots after backups, etc.

My question is - would it make sense to (after making sure I have good backups as verified by restoring to a temporary hypervisor) completely wipe the server (ESX is installed on the original RAID5 array), rebuild it as one RAID5 array across all 8 drives, then restore my VMs to it?

This would be a substantial effort on our part as there's more than 2.5 TB involved with the two VMs.  Is it worth the effort to go from one datastore spanned across 2 RAID-5 arrays to a single RAID-5 array?
Question by:mlmslex
    1 Comment
    LVL 116

    Accepted Solution

    If you are after top performance from your datastore for most workloads, if you can trade storage for performance, select RAID 10.

    If you cannot afford to do that, I would rebuild the RAID as 8 x 300GB disks in RAID 5.

    BUT, why did you not try and EXPAND the current Array, then Logical Disk....

    Also make sure you have purchased a BBWC (battery backup write cache) module for your Smart Array controller.

    and configure as 75% Write, 25% Read, this will make a difference to performance...

    also discussed here with part numbers for you to order.

    (Please note Backups do hammer and stress a datastore - what are you many VMs are being backed up at a time, and are you using software which creates a Snapshot, because performance is even worse for a VM on a snapshot disk!()

    Write Comment

    Please enter a first name

    Please enter a last name

    We will never share this with anyone.

    Featured Post

    Enabling OSINT in Activity Based Intelligence

    Activity based intelligence (ABI) requires access to all available sources of data. Recorded Future allows analysts to observe structured data on the open, deep, and dark web.

    INTRODUCTION The purpose of this document is to demonstrate the Installation and configuration, of the HP EVA 4400 SAN Storage. The name , IP and the WWN ID’s used here are not the real ones. ABOUT THE STORAGE For most of you reading this, you …
    The article will include the best Data Recovery Tools along with their Features, Capabilities, and their Download Links. Hope you’ll enjoy it and will choose the one as required by you.
    Teach the user how to rename, unmount, delete and upgrade VMFS datastores. Open vSphere Web Client: Rename VMFS and NFS datastores: Upgrade VMFS-3 volume to VMFS-5: Unmount VMFS datastore: Delete a VMFS datastore:
    Teach the user how to configure vSphere Replication and how to protect and recover VMs Open vSphere Web Client: Verify vsphere Replication is enabled: Enable vSphere Replication for a virtual machine: Verify replicated VM is created: Recover replica…

    759 members asked questions and received personalized solutions in the past 7 days.

    Join the community of 500,000 technology professionals and ask your questions.

    Join & Ask a Question

    Need Help in Real-Time?

    Connect with top rated Experts

    11 Experts available now in Live!

    Get 1:1 Help Now