Articles: Storage

Bookmark and Share

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 ]

Testbed and Methods

The following testing utilities were used:

  • IOMeter 2003.02.15
  • IOMark 0.30b14
  • FC-Test 1.0
  • PCMark 2004
  • PCMark 2005
  • PCMark Vantage

Testbed configuration:

  • ASUS P5WDG2 WS Pro mainboard
  • Intel Core 2 Duo E2160 processor
  • IBM DTLA-307015 system disk, 15GB
  • Radeon X600 graphics card
  • 1GB DDR2-800 SDRAM
  • Microsoft Windows XP Professional SP2

We installed the generic OS drivers for the drives and formatted them in FAT32 and NTFS as one partition with the default cluster size. For some tests 32GB partitions were created on the drives and formatted in FAT32 and NTFS with the default cluster size, too. The drives were connected to a Promise SATA300 TX4302 controller installed into a PCI-X slot and switched from the quiet mode (with Advanced Acoustic Management enabled) into the ordinary operating mode if necessary.

Performance in IOMark

We recorded the data-transfer graphs of the HDDs with our own IOMark tool:

The following diagram compares the read speed at the beginning and end of the partition created on each hard disk:

The Seagate 7200.4 stands out, of course. Its data-transfer rate is considerably higher thanks to the higher speed of its spindle. The speed is as high as 100MBps on the fastest cylinders at the beginning of the disk, so you can hardly now call 2.5-inch HDDs sluggish! The other 250GB-platter drives deliver similar performance, the Fujitsu MJA2 BH being somewhat faster and the Toshiba being somewhat slower than the others. The 3-platter Samsung M6 looks good in comparison with the newer products but the Hitachi 7K500 is slow.

If you take a look at the graphs, you can see that most of the HDDs produce very smooth read graphs whereas the Western Digital has a fluctuating speed. This company seems to employ a very aggressive adaptive formatting in its products to squeeze the maximum out of each particular platter.

Next we will see how the HDDs work with their buffer memory. You can see the graphs by clicking the links below and we will discuss a diagram with the top speeds.

Judging by the top speeds, the 250GB-platter models are roughly equal when it comes to working with the buffer except that the Western Digital is somewhat slower than the others and that the Fujitsu’s writing and reading speeds differ much. Hitachi has obviously replaced the electronics in the newer model as its buffer speed is much higher. We can’t say the same about Samsung’s products: the Samsung M6 seems to be just set at SATA-150 rather than SATA-300 mode.

The graphs are an interesting view, too. Hitachi’s updated electronics has a fluctuating speed of writing large data blocks into the cache: some blocks get in quickly, but others take quite a lot of time. It is even worse with the Fujitsu: the maximum speed of reading from the buffer is only achieved on very large data blocks. Otherwise, the Fujitsu’s speed is no higher than 200MBps. The new Samsung M7 is rather mediocre at writing blocks larger than 320 sectors – its speed sinks by about 25% then. Seagate’s HDDs have similar problems. Their speed lowers to about 150MBps and 100MBps when reading and writing, respectively, data blocks larger than 256 sectors. It is only the HDDs from Toshiba and Western Digital that show stable results in this test.

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 ]


Comments currently: 0

Add your Comment