The Physical To Virtual Switch

The Physical To Virtual Switch

Not everyone has already made the leap from physical devices for each application. I’m about to embark on a physical to virtual project with one of my consulting clients. Over the next couple of days I’m going to explore some of the issues we think will need to be resolved in any P2V project.

Many businesses take a passive approach to planning their IT. Rather than setting out a specific strategy, they start with the least possible investment in technology and slowly add systems on an “as needs” basis. The problem is that after a few years, they’re left with systems that aren’t properly secured with ad hoc backups and poor utilisation of storage and processing capacity.

When that happens, or looks like it’s going to happen, it’s time to take stock and evaluate what to do next. That’s the position I’m in at the moment with a client. Several applications are running on separate desktop-class computers and the existing server platform used Microsoft Small Business Server – and the license limitations of that are now being hit.

So, what should I advise my client? Here’s what I’m proposing.

We have about 10 applications running on a mix of Windows and Linux. Some of those applications collect in excess of 2GB of data each day. There’s a need to get smarter about storage management and backups as well.

The proposed solution is based on a pair of physical servers running all the applications in a virtualised environment. There will be a storage appliance that communicates with the physical hosts. Data will be backed up to disk – the plan is to redeploy an existing server with some additional disk to make it the first tier of a backup solution. The disk-based system will be backed up to tape for offsite storage.

All of this is a reasonably non-controversial approach as far as our research and analysis reveals.

Over the next couple of days, we’ll discuss how we chose a virtualisation platform and the journey we took with the business in getting them to see the business benefits – and not the technical reasons – of this solution.


  • Your solution of 2 hypervisors with shared storage seems reasonable and you may even be able to consider the use of local storage technologies with these hypervisors if the IOPS workloads of the applications aren’t too heavy.

    One of the business benefits I would be really trying to push home is the opportunity to remove the offsite tape from the equation. Tapes need manual intervention and need to be rotated back from your offsite provider (or persons home) to ensure data is retained.

    Using cloud storage providers for long term backups seems like a no-brainer in my mind (Glacier etc) however people are still reluctant. To get around this reluctance I recently deployed a pair of Synology NAS devices that automatically RSYNC (disk to disk) to each other nightly from the small business to the house of the director. This could of course be another site/office if the business is that large. Some management need to ‘see’ blinking lights and their investment in technology for themselves.

    At the end of the day you are consulting and need to provide the pros-cons to them and let them decide what way they want to go.

    I really look forward to reading how you go ‘changing minds’ over the next few days!



  • For a cheapish DR/BC solution, Physical host with DAS, Veeam and replicated the machines. I use this for a lot of my SMB and this gives them backups as well as the ability to fail over.

Comments are closed.

Log in to comment on this story!