Usage

I have a rather large desk, and I don't think I could get a much bigger one with getting an L-shaped desk or switching to a folding table. As such, I really don't have room for an external drive in plain site, so I've opted to sit it behind my monitors. The desk itself is hardwood. After plugging the drive in with the eSATA cable and powering it on, I could feel my desk vibrating any time I laid my arms against it. Worse, if all my drawers are closed, the vibration becomes slightly audible! If I'm going to keep this drive where it is, I'm going to have to find some sort of rubber or foam mat to put under it to prevent these vibrations.

eSATA is only an extension of internal SATA, where on the cable has been modified to allow for external use. This means that the drive will work fine in Linux or Windows provided the SATA controller is supported. Similarly, the enclosure can also use a slower USB connection which is supported by both Linux and Windows as well.



For testing, we'll be using a Rosewill-branded SATA controller. This controller is only of the SATA I variety, and therefore we're going to test a SATA I controller. SATA I has about half the theoretical bandwidth that SATA II has, but we shouldn't see a major gap in performance. The tests will be performed on our Intel Quad Core Q6600 machine with 2GB of RAM.

Testing & Benchmarks

In Windows, we've decided to test two ways: using synthetic benchmarks with Sandra, and the real-life benchmark of copying files between drives. In our real-life benchmark, we'll be transferring a 700MB file from a Western Digital Raptor drive to our Vantec enclosure's drive, and then back. Let's start with the real-life benchmark:

Direction	Connection	File Size	Time	Speed
To enclosure (write)	SATA	700MB	15.0s	46.67MB/s
To enclosure (write)	eSATA	700MB	15.1s	46.36MB/s
To enclosure (write)	USB	700MB	26.2	26.72MB/s
From enclosure (read)	SATA	700MB	15.0s	46.67MB/s
From enclosure (read)	eSATA	700MB	15.0s	46.67MB/s
From enclosure (read)	USB	700MB	20.6s	33.98MB/s

As it can be surmised from our table, there's little difference on whether you use the SATA drive through an internal SATA connection or external SATA connection, as the speed remains constant. However, when you switch to the USB connection (maybe for a laptop, or lack of an eSATA port), you will see a drop in performance. I was surprised at the equal read/write speeds for the eSATA and SATA connections, it seems a little odd. Writes should be much slower.

On the synthetic side of things, Sandra reported a Drive Index of 48MB/s for both the eSATA and SATA connections of this drive. Although this is more of a theoretical read speed, it doesn't differ much from what we saw in the table above. Using a USB connection, Sandra reported a drive index of 32MB/s. Again, this does not stray far from what we saw in the above table.

Since we're only looking at the enclosure itself and not the card, we feel it's safe to say that the enclosure does not hinder drive performance. That is, as long as the eSATA connection is utilized and not the USB.

Our final test involved the heat given off by the drive. As the enclosure does not have a fan, we were worried about the heat that may build up in the enclosure if the drive was used for an extended period.

I wrote a simple Linux shell script to continuously copy our 700MB file to the drive 100 times. In total, this took about a half hour to complete. When I touched the enclosure anytime during the test, the enclosure still felt cool to the touch. Imagine the power of an aluminum enclosure! Of course, I'm sure some drives with larger caches or faster RPM will warm the enclosure up much more.