Computer Geeks eSATA Enclosure HE-2521B
Author: Zahn Funk
Editor: Shawn Knight
Date: 02-20-2010
Provided by: Computer Geeks

Using the provided screws the drive is mounted to the PCB through these slotted holes. Although a Toshiba 80GB 5400RPM drive is pictured, this was later exchanged for a Hitachi 100GB 7200RPM drive for better performance analysis. The tray is slid back into the housing and the two screws replaced on the sides. A nice vinyl sleeve is included that fits tightly over the case and will undoubtedly help protect the finish, plus it covers up the lettering as well. It may also hold the heat a little more but this did not become an issue during testing as temperatures never exceeded safe thresholds.

A powered eSATA cable is included which has a thick rubber coating and feels very heavy duty. This connector will also plug into a standard non-powered eSATA port however in this case an alternate power source must be used. Since no AC adapter is included with the product, the USB cable must be used to provide power for the drive and electronics. Having two cables connecting the drive to the laptop is still preferable to being tied to an AC adapter, although it will of course use battery life more quickly. The power/activity LED glows green when the drive is getting power, and flashes orange when read/write activity is present.

I'm using a Lenovo T500 laptop and for this first test I connected the USB cable by itself to measure throughput via USB 2.0 speeds. As you can see, maximum read speed is capped right around 28-29MB/s, far short of the theoretical 60MB/s limit but not unheard of for an external USB drive. I believe the best I've ever seen USB 2.0 perform in a long, sustained read was around 35MB/s. The performance shortfall may be affected by the SATA to USB bridge chip or the USB controller in the host system, or possibly even a little of both.

For the second series of tests I'm using a Vantec eSATA ExpressCard which does not provide power over the eSATA, so the USB cable must also remain connected to power the drive. Without the flat rate limit of USB we're now seeing maximum performance of the drive itself, which averages around 40MB/s sustained read. Write speeds are not much farther behind, although throughput drops off significantly for smaller, random read/writes. Still, the exercise shows that using the eSATA interface will allow for the full bandwidth available to max out a traditional hard drive, and should provide at least as good of speeds for a solid state drive as well. Also using an eSATA connection appears to utilize less CPU resources as well, freeing up processing power for more important things than managing bandwidth.

  Sponsors Links

  Sponsors Links