Server Virtualization is a great way to enhance your disaster recovery strategy. Server Virtualization adds agility not only in managing your day to day server environment but it also adds agility in restoring your servers in case of a problem.
When restoring a server to new hardware it is really nice when you can restore it to the exact same hardware. For example if the original hardware has a disk controller for an array of disks then it is much easier if the new hardware has that same setup. After a full restore, new hardware with a different controller card, won’t be able to access the disk storage after your reboot. This can be very frustrating and add hours to each server restore. You will have similar problems with other hardware such as network cards. Anyone who has done a test restore or a real disaster recovery will tell you this is a huge pain.
Doing server restores to new hardware is one area where virtualization really shines for a disaster recovery. First, you will need to restore your virtualization host servers. You will run into the same driver issues here but once the hosts are up all of you VMs just work. Because the VMs are running under the virtual environment the virtual hardware is all the same so there are no driver updates needed. This enables a much faster recover and a lot less gnashing of your teeth. This benefit is the same for Citrix XEN, Microsoft Hyper-v, and VMware ESX.
There are other areas where server virtualization can help improve you DR recovery time. One example of this is using a virtualization cluster which can start your VMs automatically on another host server in case of a server hardware failure. This addresses just a local server issue and not a catastrophic failure at the Data Center. Another virtualization technology that can provide faster recovery time from a catastrophic datacenter failure is a technology like VMware’s Site Recovery Manager or Microsoft’s Hyper-v Replica. This will keep a copy of the running VM at another site. It takes just minutes to have the server up and running at the backup location. Depending on the risk you are willing to accept this may be all the DR protection you need for a server. Others you may want to add other backups either online or offline.
As Jennelle wrote about in her post yesterday, the first thing to do in your DR planning is rate the needs of your services. How quickly do they need to be up and running? How much data can be lost, if any. It would be nice to set everything to the highest level of priority on both speed to restore and importance of data retention. Unfortunately, it is cost prohibitive to offer the highest levels of protection to all data. The other thing to keep in mind is the business that you are in. For example, a military is more likely to have a disaster at two geographically dispersed datacenters on the same day than say a bicycle manufacturer. Once you understand the different needs for your services and data you can match those needs to the available tools. There are are a lot of tools available that can provide for various levels of data recovery and speed to restore. Virtualization adds a lot of options for your tool belt.
Below is a list of a few tools used in conjunction with virtualization that are useful as you look to balance costs and needs in regards to your disaster recover and business continuity plans.
- Windows Server Backup (supports Hyper-v for more details go here)
- Hyper-v Replica
- Veem – Backup
- Vision Solutions – Double Take
- System Center – Data Protection Manager
In other post of this series we will take a deeper look at specific tools like Hyper-v Replica and explain how it works and step you through implementation.
This is post part of a 15 part series on Disaster Recovery and Business Continuity planning by the US based Microsoft IT Evangelists. For the full list of articles in this series see the intro post located here: http://mythoughtsonit.com/2014/02/intro-to-series-disaster-recovery-planning-for-i-t-pros/