4x Samsung SpinPoint T166 500 GB (HD501LJ, Serial ATA II, 7200 RPM, 16 MB)
External Disks:
Seagate Barracuda ST350063 in USB 3.0 enclosure
Firmware:
3.6.1 Build 0302T
Ethernet Switch:
TL-SG3216 16-port Gigabit managed switch (LACP and Jumbo Frames support)
Ethernet Cabling:
CAT 6e, 2 m
Methodology
We use three different programs to evaluate the performance of the NAS. The first is Intel's NAS Performance Toolkit. Intel was kind enough to build a NAS performance toolkit and not only release it for free to the public, but also provide its source code. The only problem of this toolkit is that if you use a client PC with more than 2 GB of memory then the results of two tests ("HD Video Record" and "File Copy to NAS") are heavily affected, since they actually measure the client's RAM buffer speed and not the network speed. We are not sure also if the results of the rest tests are affected (and if yes in what degree) when a client with more than 2 GB memory is used. During all tests with this toolkit we exploit its batch run function, which repeats the selected tests for five turns and uses the average readings as final results.
The second program is a custom made program which performs ten basic file transfer tests and for each measures the average MB/s speed. To extract as accurate as possible results we run all selected tests ten times and in the end we use the average as the final result.
The third program we use in our test sessions is ATTO, a well known program for storage benchmarks. In order to use ATTO for benchmarking we are forced to map a shared folder of the NAS to a local drive, since ATTO cannot directly access network devices.