Initial Performance Tests: Conclusions
As to the eSATA experiencing slower backup times between the three tested devices, I would have to think that the hardware and/or backup software on the sending side may possibly add some extra overhead when sending data out to the eSATA device. From the results generated by OpenSourceMark and TRMark, I would expect a faster backup going to the eSATA chassis drives. If you think about it, the first two tests are just stripped down file copying processes that don't require calls from a management tool to start up and function. I would have to consider that the backup software may have possibly added a degree of latency, preventing the eSATA to reach its expected results. The problem may even go deeper than that since the IDE and USB backup tests should have suffered like the eSATA test, but instead ran at 41 and 46 seconds faster, respectively.
Standard Performance Benchmarks
I ran the following benchmark tests on a select group of the 3.5 inch Seagate Barracuda 7200.10 750 GB SATA drives in the High-Rely eSATA chassis. I used three benchmarking programs:
It was interesting to see how the same model SATA drives performed against each other in the same external chassis. I tested four of the seven drives that Highly Reliable Systems sent me. One particular drive I tested, the I: drive, definitely stood out when I ran the H2BenchW and IOMeter testing. It performed best in the h2benchW sequential transfer tests, especially on writes. However, it did not perform as well as the other three drives on the IOMeter IO Operations benchmarks. I'll be digging into these unusual test results with Highly Reliable Systems. I'll let you know what we come up with.