Articles: Storage
 

Bookmark and Share

(6) 
Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 ]

Testbed and Methods

The following testing utilities were used:

  • IOMark 0.30b14
  • IOMeter 2003.02.15
  • FC-Test 1.0
  • PCMark 2004
  • PCMark 2005
  • PCMark Vantage
  • Raxco Perfect Disk 8.0
  • WinRAR 3.60

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 (and Windows Vista for PCMark Vantage)

The HDDs were tested with the generic OS drivers. We formatted them in FAT32 and NTFS as one partition with the default cluster size. The HDDs were connected to a mainboard port and worked with enabled AHCI.

As for the WD20EARS with 4KB sectors, we already know how an unaligned HDD with such long physical sectors might perform because we tested the 1-terabyte model. Therefore in this test session disk partitions are aligned for this HDD with WD Align in each test. When benchmarking it in IOMeter’s tests that involve writing, we have two test passes: one is standard (we call it “unaligned”) and the other is “aligned” (the LBA addresses of requests are multiplies of 4 KB and match the HDD’s physical structure ideally). The difference is quite noticeable as you will see. We do not align those tests that do not require that, e.g. the test of sequential reading (aligning 2KB requests to 4KB sectors does not make sense) or multithreaded operations (the request size is 64KB).

 
Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 ]

Discussion

Comments currently: 6
Discussion started: 04/19/10 07:58:21 PM
Latest comment: 10/27/11 05:48:36 AM

View comments

Add your Comment