Revisit your DR strategy

Backup is one of those “non-sexy” things that often gets ignored while companies focus on the latest shiny bits.  After all, it’s very much like insurance; we all have to have it but we pretty much ignore it until the next time we have to pay the premium.  That’s a pity but it is human nature.  I’d like to suggest that you take a few moments to revisit your backup strategy and make sure that you are covering the bases.  And yes, I know, this is a topic I have banged on about more than once before …

With the movement to more and more virtualised workloads it makes sense to explore the options available to backup at the host level rather than at the file level because there are some truly awesome tools out there.  Virtualisation tends to lead to “server sprawl” and ever increasing amounts of data to back up.  A lot of that data may be duplicated and a lot may be “static” an unchanging.  It makes sense, then, to have technologies in place that can perform some level of de-duplication and manage static data “intelligently” during backup.  At the high-end of the marketplace (eg the enterprise level) there are all sorts of options in both hardware and software.  Surprisingly, though, there are also options that can be utilised by smaller companies even those that might only have a “single host” virtual environment.

Veeam (of course!) is at the top of my list but there are also products like Dell AppAssure and others out there that offer many similar capabilities.  Dell AppAssure, as an example, offers “Veeam-like” backup capabilities for both virtualised servers as well as physical servers which could be a large plus if you have an environment that has lots of both types of servers.

My point is not to have you start comparing the relative merits of products like Veeam or AppAssure but rather to take a step back and cast a cold, hard look across your backup and DR design/strategy and see if it still “fits” with your requirements.  Frankly, if your design is more than a couple of years old you might well be shocked at what you are missing.  The backup/DR world is moving at an incredible pace which is a good thing considering how fast things have been changing in the virtualised world (Hyper-V 3, anyone???  ESXi 5.1???).  With appropriate backup/DR tools in place you can have almost immediate recovery from a server or host failure, far far faster than you could ever hope for with traditional file-based backup.  In fact, we see such super-quick recovery capabilities as one of the compelling drivers for going to a virtualised environment above and beyond all of the other drivers.

The other thing that comes along with the backup and quick recovery capabilities is being able to leverage replication through a tool other than that provided by your virtualisation vendor.  Replication in VMware usually requires additional licensing above and beyond what small firms will buy (think VMware Essentials).  And while the latest Hyper-V 3 does have some awesome replication capabilities there may be reasons why you might want to handle replication separately.  Veeam (as an example) has some interesting abilities to handle both replication of backed up data to a remote site as well as control actual replication of virtual machines through a process that is “external” to the virtualisation vendor’s tools.  This gives you some options that you might not otherwise have and it’s always nice to have options.

Is there “one correct way” to create a backup/DR design?  Of course not!  That’s why the market provides so many tools and options.  But only an extreme risk-taker (or a technical person living in a vacuum) would assume once a backup/DR design is in place that everything now stays “static”.  Backup requirements (and abilities), like everything else in the tech world, is fluid.  Take some time and do your review, specially so if you have moved on to a virtualised platform and have not updated your backup/DR design.  The payback on the effort is worth it!