Technology can be a wonderful thing, can’t it? It wasn’t too long ago that having any kind of off-site disaster recovery solution in your company meant that you were a member of the Fortune 500. Well that’s not true any longer. In fact, this technology is so affordable now that virtually any size company can implement one of several possible disaster recovery solutions and protect themselves from catastrophe. So why is that? Three key things… the widespread acceptance of server virtualization, the availability of inexpensive high-speed internet connectivity, and new low-cost disaster recovery software solutions tailored to the virtual world.

Let’s take a look into the past and review where we have come from. There have been three phases in the evolution of this function:

The Dark Ages. Starting in the 1960s and progressing through the early 2000s, complete or mostly complete duplicate data centers needed to be maintained by the company itself or in conjunction with a disaster recovery backup specialty company such as Sungard. In this scenario, the company would need mainframe or mini-computer resources of the type they were using available in the backup location, and they would transport backup tapes to that site when it needed to activated. Pretty slow and painful, as I’m sure many of you can remember and attest to. Oh and there’s one other problem, it was EXPENSIVE.

Virtualization--Phase 1. So about 10 years ago, the trend to consolidating most functions onto Windows servers was well under way, and VMware started their push to server virtualization prominence. Server virtualization was (and is) great... better utilization per host server, VMotion for dynamic load balancing, and best of all, VM portability to run transparently on any host that has access to the appropriate data storage. Fantastic.  Simultaneously, the Internet was exploding and high-speed connections were becoming available at reasonable costs.

At the same time, and as a direct result of virtualization and cheap high-speed connections, Co-Location and Managed Services Data Centers started to show up everywhere. Web servers were the first function that were commonly outsourced, then Salesforce automation, email, and more. It has now become accepted and even expected that outsourcing will be used by most companies in some form or another. This confirms that acceptance issues regarding storing company data outside the corporate environment, is for the most part, a thing of the past.

VMware then took the next step in offering products like Site Recovery Manager. This allowed companies to manage the replication of their data to a company owned or third party remote site, establish a runbook to automatically start-up the entire environment in the proper order, and to make it available in very short order to company users. With proper continuous data replication so that the datastore is constantly in sync, even large environment can be transitioned from primary to backup in 30 minutes or less.

For the last few years while this technology was available, we would have this discussion with companies wanting to implement a structure like this. Their executives had heard that this could be done and they were all for this type of solution and the insurance and security that it brought… until they saw the price tag. Then there was a lot of back peddling. This was light years better than the process in the dark ages, but it was still too expensive for most SMB companies to attempt.

Virtualization--Phase 2. So here we are today and a few things have changed to reduce the price of entry. First of all, CO-LO providers are offering VMware hosts for use during an emergency. They are doing this on a dedicated basis where the host are for your use only at a certain price, or on a shared basis where that capacity might be sold to 3 different companies banking on the fact that not all companies will need it at the same time. This service is obviously at a lower cost than the dedicated service.

Second, there is much more cost-effective software available that is built from the ground up to work with virtual rather than physical servers. Examples of this are from companies like PHD Virtual. This software effectively handles the backup of all types of Virtual servers, the data replication to a remote site, and the runbook functionality required to transition quickly and effectively to your remote computing resources. All of this for a fraction of what has been the norm for Site Recovery Manager and some form of SAN-to-SAN replication.

Third, this new software also addresses the costly manner in which real-time data replication was accomplished. As alluded to above, the best way to do this until recently was SAN-to-SAN replication, meaning that the same brand SAN needed to be at both sites. This could be a significant added expense with the traditional SAN vendors (like EMC and NetApp), though Dell EqualLogic has done a great job of making this affordable at the low end. The good news is that with the latest generation of VM backup and data recovery software, the replication is handled directly through software from any source to any target, which save money and provides ultimate flexibility, especially when rent resources from a CO-LO company need to target data onto their SAN which is likely a different brand from your storage.

In conclusion, virtualization—phase 2 is here. The bar is now significantly lower in terms of cost to implement a fully featured solution of this type, which makes it available to smaller and smaller companies, and this kind of protection is the best insurance your company can have in today’s non-stop world.