Economics of Video Surveillance RAID

While valued by security managers for years, use of RAID for video surveillance systems has been low. With many organizations migrating to software only solutions, the RAID decision has new challenges.

UPDATE February 2012: This article is out of date has been replaced by our 2012 Surveillance RAID Usage Statistics/Analysis.

Today, Milestone Systems published an article on "To RAID or not to RAID?". It's well written and worth reading but two immediate concerns came to mind:

• The Key Element is the Cost Difference
• For RAID, software only systems seem to be worse than older appliances.

Most security customers value the storage redundancy of RAID but have not been able to justify the increased cost. In this respect, RAID is like megapixel cameras. Most everyone values the higher resolution of megapixel but only a minority of security managers can currently justify the additional expense. Therefore, the key element in the decision is how much more RAID costs.

The RAID premium has continued to decrease over the years. 5 years ago, using RAID for Intellex systems increased the cost of a 16 channel appliance by 50% (from roughly $10k to $15k. Today, using RAID on Intellex systems increasing the cost of the equivalent appliance by less than 25% (from roughly $9k to $11k).

Last year, 3VR offered RAID standard on its entry level appliances, effectively eliminating any premium for RAID. At that point, choosing RAID becomes a no-brainer. Expect this trend to continue as PC motherboards increasingly support software based RAID.

For appliances, at least, the RAID premium has dropped under $2k for a 16 channel system and is approaching $0. As it does, value is generated as the cost for RAID drops below the amount of savings generated from not losing evidence and saving systems from going off-line due to a hard drive failure.

Software Only

The other interesting element of this article is how much more costly and complicated selecting and setting up RAID is for a software only solution than an appliance.

This is a regression or a step back from what security managers and integrators have been experiencing for years. With appliances, adding in RAID is like super-sizing your meal or ordering an iPod. You simply select what you want and the product is delivered to you, fully integrated and ready to use.

As Roger describes in his article, "RAID technology is not particularly difficult to configure, but it can be somewhat confusing. Each product has a different interface, and most of them rely on a BIOS level utility program to initialize the disks and build arrays. RAID is not for the faint-of-heart."

In business terms, this increases cost by requiring more time spent on set-up and using a more skilled technician to accomplish the setup. Ironically, this could reduce the value of RAID to security managers. These complexities will likely spur the adoption of products by Intransa or Pivot3 as organizations seek to simplify and reduce the time and risk of manually setting up RAID arrays.

It would be interesting to see more details on the relative costs of using RAID in software only versus appliances. In this way, integrators and security managers could make better decisions about the value generated from using RAID.