|Patriot Gauntlet PCGT25S USB3 Drive Enclosure|
|Reviews - Featured Reviews: Storage|
|Written by David Ramsey|
|Monday, 23 August 2010|
Page 4 of 5
Testing & Results
USB 3.0, aka "SuperSpeed USB", was introduced at the 2010 Consumer Electronics Show in Las Vegas. At almost 5 Mb/s, its raw data rate is ten times that of USB 2.0, which should result in much faster transfers of data...as long as you're connected to a USB 3.0 device. The Patriot Gauntlet SuperSpeed USB Edition comes with a USB 3.0 cable; while this cable is physically and electrically different from a USB 2.0 cable, it's been designed so that it will plug into USB 2.0 connections, although this will obviously result in the connected device running at USB 2.0 speeds.
USB 2.0 was plenty fast when it was introduced in late 2001, since flash memory, laptop drives, and the like were all much slower than they are today. But newer drives and devices easily overwhelm the bandwidth available to USB 2.0 and can really benefit from SuperSpeed USB's faster data rate. For this review I installed an OCZ Vertex Turbo SSD into the Gauntlet and ran benchmarks comparing the USB 2.0 and USB 3.0 rates with native SATA.
Like most USB 3.0-supporting Intel X58 motherboards, the Rampage III Extreme uses NEC's ubiquitous D720200F1 USB 3.0 controller to enable its two USB 3.0 ports. Different benchmarks reveal different things, and I tested the Patriot PCGT25S with the ATTO Disk Benchmark, CrystalDiskMark, AS SSD, and the disk benchmark built into Everest Ultimate Edition v5.50. The results were...interesting.
The ATTO Disk Benchmark is a free tool that, while old, still offers a good look into basic storage system performance. It measures interface transfer rates at various intervals for a user-specified length and then reports read and write speeds for these spot-tests. Our benchmarks are conducted with a queue depth of 4, overlapped I/O, and transfer sizes from 0.5 to 8192K with a total length of 256MB.
As you can see, USB 3.0 offers a tremendous increase in performance over USB 2.0, with reads that are 5.8 times faster than USB 2.0, and writes that are 6 times faster. However, it's still not as fast as native SATA, which turns in scores that are 28% better for reads than USB 3.0, and writes that are 9.7% better.
The disk benchmark built into Everest Ultimate is different from the ATTO program: it performs straight linear reads and writes across the entire disk, ignoring any partitions or formatting on the device being tested (which means the write test will wipe your disk, so be cognizant of this). Benchmark Reviews sets the block size to 1MB for this test.
The performance different between USB 2.0 and USB 3.0 as reported by Everest are similar to the ATTO results, with USB 3.0 providing reads that are 5.8 and writes that are 6.3 times faster than USB 2.0, but native SATA still turns in performance that's 23% better than USB 3.0 for reads and a substantial 35% better for writes.
But storage system performance isn't just about raw read and write speeds, it's also about how rapidly the system can respond to multiple disparate I/O requests, and this is measured in "IOPS", or "input/output operations per second." This is an important aspect of storage subsystem performance that even many enthusiasts overlook; although it's not obvious, in a modern system there may be a dozen or more background processes accessing the disk at any one time. SSDs typically perform much better than hard disks in this regard since they don't have to contend with rotational and head-movement latency.
CrystalDiskMark 3.0 is a file transfer and operational bandwidth benchmark tool from Crystal Dew World that offers performance transfer speed results using sequential, 512KB random, and 4KB random samples. For our test results chart below, the 4KB 32-Queue Depth read and write performance was measured using a 1000MB space. Rather than measuring simple read and write bandwidth, this test stresses the attached storage system with multiple small I/O requests.
USB 3.0 turns in reads that are 180% faster than USB 2.0, but still only a fraction of what SATA can provide. The performance difference between USB 3.0 and SATA is much larger than we saw in the previous tests.
The last test is Alex Schepeljanski's AS SSD Senchmark. The AS SSD Benchmark tests sequential read and write speeds, input/output operational performance, and response times. Because this software receives frequent updates, Benchmark Reviews recommends that you compare results only within the same version family. I used the 4K-64 thread test to compare IOPS performance.
The results are quite similar to those reported by CrystalDiskMark: USB 3.0 performs better than USB 2.0, but both are dwarfed by the performance of a native SATA connection.