Solid State Disk (SSD) is often the solution to drive up performance of a particular application, increasing response time to users and thereby increasing revenue or productivity. There can also be significant CAPEX savings by implementing SSD, what is often missing from the discussion is the operational or OPEX savings that comes from implementing SSDs.The CAPEX savings from implementing SSDs, as we discussed in our article "SSD's are Cost Effective Now", generally come from reducing the number of mechanical drives that are being applied to an application. Given the right queue depth (number of disk I/O requests) the more drives you have in the RAID group the better performance of the application becomes. As the queue empties because there is a high enough drive count, performance no longer improves as drives are added. The only remaining move for a storage manager is to improve response time. This is first done by implementing faster drives but once you are at 15k RPM speed then more of a drastic measure must be taken. As we discuss in a prior Information Week entry, you have to short stroke the drive.

On traditional storage systems implementing and creating all of these work-arounds for drive performance can be very expensive, which is how SSDs can quickly become a cost effective alternative. One SSD can eliminate all of these extra drives, wasted space and power consumption while delivering better performance. It is the operational overhead of a high drive count RAID group that is often left out of the discussion and is one of the bigger ROI's that a SSD implementation can deliver. The time it takes to manage and a 300 drive RAID group with short stroked drives is daunting, especially when compared to managing just one SSD device.

SSD OPEX savings goes further however. I've seen storage managers spend weeks trying to design the perfect LUN layout for a new application or for an old application that is underperforming. For example one storage manager I spoke with was trying to decide how to configure his SAN and LUN design for optimal performance on a relatively large Exchange Server. After telling me all of his ideas, most of which were sound and well thought out, I suggested putting the whole thing on SSD. First reaction? Too expensive. In reality once the configurations were compared, the SSD system was about 10% more expensive but then he would be done. There was no tuning need now or later. Just put it on the SSD and it flew. That 10% more saved him weeks of planning plus who knows how much time fine tuning the storage configuration as the environment continued to grow. That's time he could easily use elsewhere and he did.

Track us on Twitter: http://twitter.com/storageswiss

Subscribe to our RSS feed.

George Crump is founder of Storage Switzerland, an analyst firm focused on the virtualization and storage marketplaces. It provides strategic consulting and analysis to storage users, suppliers, and integrators. An industry veteran of more than 25 years, Crump has held engineering and sales positions at various IT industry manufacturers and integrators. Prior to Storage Switzerland, he was CTO at one of the nation's largest integrators.