Aug
5

An old cowboy once said, “Redundancy equals reliability.” But how do you get it for Windows servers on the cheap?

An old cowboy once said, “Redundancy equals reliability.” But how do you get it for Windows servers on the cheap?

August 5
By

There’s simply no better way to increase system reliability than redundancy.  But, in a physical world, redundancy is extremely expensive.  Exponentially expensive.  Want more reliability for your Windows Server systems?  Sure, totally doable, but it will cost you.  A lot.  Mirrored systems.  Clustered systems.  Synchronous communication links.  All pretty cool – and a pretty penny.Creating a redundant system

Enter virtualization.  Being able to have a physical or virtual server mirrored to a virtual server or virtual cluster is also pretty cool, but if you start to look at the underlying infrastructure needed to have live standby/failover servers and/or virtual clusters, wow– it’s still a lot of cash.  And it gets complicated quickly as well.  Larger, resource-rich enterprises with zero tolerance for downtime and data loss pay these premiums because they must, but they’re happy these extreme high-availability (HA) solutions are available.  But that’s not what most businesses need.  So while virtualization and advanced HA technologies are a key piece for providing system reliability via redundancy, it’s not the entire answer for most of us.  And even continuous data protection (CDP) is very costly in terms of computing overhead and complexity.Node failover scenario guide

I’m the grandson of a frontier cowboy, Joe Singleton, and it’s just part of my nature to solve a problem with a simple, inexpensive fix that gets the job done rather than a complex, expensive one that get the job done perfectly.  So, cowboys, how do we get to the point of using redundancy to enable very quick Windows Server system recovery (because downtime is expensive too) with minimal data loss at a price point that’s friendly?

For Windows Server systems, continuous snapshot backups coupled with very quick recovery techniques deliver high reliability at such a compelling price point that even Grandpa Joe would have given his approval.  It’s a simple, inexpensive fix that gets the job done.

Think about it — let’s stage a redundant, virtual recovery server, instead of  having a live failover server or cluster, to reduce system resource requirements and complexity. Let’s leverage an open-source, free, simple hypervisor for quick recovery to reduce cost and complexity. Let’s have recovery points of every 15 minutes instead of CDP to reduce processing overhead and complexity.  Let’s leverage lesser hardware to have a quick recovery solution with a little less performance than a production server but which will get us up and running and cost a lot less than fully redundant hardware.  Let’s choose manual failover over automated failover to reduce cost and complexity.  Let’s aim for near-enterprise high-availability and near-CDP using a simple, efficient, cost-conscious method to get the job done.

Enter the StorageCraft Recover-Ability Solution which delivers highly efficient incremental backup snapshots every 15 minutes coupled with very quick recovery/restore features VirtualBoot and HeadStart Restore and much more.The storagecraft recover-ability solution

pack-horse-and-cowboy

So, did my Grandpa Joe really say “Redundancy Equals Reliability”? Well, it went more like “If you’re making a long trip through rough country, taking a pack horse along is a good way to go. If one horse up and dies on you, you can still get yourself out of there.  When it’s rough, two horses is better than one.”  RIP Grandpa Joe.