Monday, July 25, 2011

George Crump, Senior Analyst

After installation our first test was to gather some simple baseline benchmarks. In the first workstation we installed the 240GB Deneva 2 in was an Intel quad core i7 based Mac with 8GB’s of RAM. This system certainly has the horsepower to push the drive and it also runs some very performance demanding applications that render videos and convert audio. For benchmarking we use xBench which among other tests gives us disk performance reports.

There is often debate about whether or not a workstation should use solid state storage. Typically a single workstation cannot generate enough parallelism to push a solid state disk to its limits. This may be true but that does not mean that a solid state disk can not provide a noticeable performance improvement to the user, especially if their need is more performance driven than it is capacity driven.

For example this workstation houses our projects database which is performance demanding but not large. The movies and podcasts that are created are large but after creation they are pushed to a NAS for final storage. If you look at the capabilities of the system by far the slowest component is the mechanical hard drive that it came with. We have plenty of network attached capacity, what we needed for this system is blazing fast performance. In fact the 240GB Deneva 2 has plenty of capacity for our needs.

We used the original hard drive to perform our base comparison. The Toshiba hard disk drive that was in the system was tested using Xbench. While not a fair comparison at all, it certainly provided a glimpse into the performance difference we were about to see.  From the screen shot below you can see that this was a very average drive holding back a high performance system.

The next test was to install the Deneva 2 and test it. While our configuration is no where near capable of getting maximum performance out of the drive the results were impressive and noticeable. Needless to say it far surpassed the performance of the internal HDD. While it may seem like an unfair test it is exactly the performance difference that a user will “feel” when they make a similar switch.

More impressive is that it has scored the best of any internal SSD we have tested thus far in the lab. In most cases the Deneva is 2-3 times faster than any drive we have tested so far. Again, beyond the benchmarks the system feels snappier than using our past test of other pure SSD systems.

Benchmarks are fine, but you buy one of these drives because it makes you more productive and I am pleased to report that  has been our findings. The most astonishing was the first time we powered up the system with the new drive installed. It was so fast that I thought we didn’t actually select reboot, so I rebooted again and watched the screen. Within seconds our system was up and ready. More importantly than fast boots we find that look ups into our project database are now instant and the encoding of video and audio have seen a significant improvement in performance. In short we can get more work done in less time.

Consider the first test a success. Plug and play installation, impressive benchmark scores and a user noticeable improvement in performance, leading to greater productivity. As is the Storage Switzerland style we will keep pounding on the drive and report back on how it is holding up on performance and reliability. The second drive we will install into our servers to measure the performance improvement and relative value of the OCZ Deneva 2 in more server class workloads.