Testbed and Methods
The following testing utilities were used:
- IOMeter 2003.02.15
- WinBench 99 2.0
- FC-Test 1.0
- PCMark 2004
- Albatron PX865PE Pro II mainboard
- Intel Pentium 4 2.4GHz CPU
- IBM DTLA-307015 system disk, 15GB
- Radeon 7000 32MB graphics card
- 256MB DDR2-533 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 and formatted in FAT32 and NTFS with the default cluster size. The HDDs were attached to a Promise SATA150 TX2 Plus controller that supported NCQ and were switched from the quiet (with the Advanced Acoustic Management enabled) into ordinary operation mode.
Performance in Intel IOMeter
Sequential Read & Write Patterns
The HDD is receiving a stream of read and write requests with a request queue depth of 4. The size of the requested data block is changed each minute, so that you could see how the drive’s sequential read/write speed depends on the size of the data block.
That’s a good illustration of the influence of the spindle rotation speed on performance: the MXH2250BT is about 15MB/s slower than the others in terms of maximum read speed. The 5400rpm HDDs have similar results, achieving their top speed of 55MB/s on 16KB and larger data blocks. The Fujitsu MHY2250BH has a somewhat lower top speed, yet it is far faster than the MXH2250BT, too.
Note the superb performance of the WD2500EVS on small data blocks: it is 3-5MB/s ahead of its closest pursuer Hitachi on all blocks smaller than 16KB.
The HDDs are different at writing. Of course, the Fujitsu MXH2250BT has a low speed as might have been predicted, but the Samsung HM250JI has a surprisingly low top speed at writing, being 10MB/s slower than the leaders.
The top speed is achieved on different data blocks, too. The HDDs from Samsung and Western Digital reach it as soon as 8KB blocks while the HTS542525K9SA00 and MHX2250BT do it on 16KB blocks. The MHY2250BH accelerates to its top speed on 32KB blocks only.
The HDD from Western Digital is unrivalled when writing small data chunks. Its performance is about two times that of the Samsung and Hitachi. The HDDs from Fujitsu are only half as fast as the latter two.
In this test IOMeter is sending a stream of requests to read and write 512-byte data blocks with a request queue of 1 for 10 minutes. The total number of requests processed by the HDD is over 60 thousand, so we get a sustained response time that doesn’t depend on the HDD’s buffer size. The results are sorted by read response time.
It’s easy to identify the 4200rpm model even if you erase the labels from the diagram: the MHX2250BT is about 15% slower than the others in terms of both read and write response. The high write response of the Samsung should be noted among the others. Samsung’s 160GB model had a rather poor result in this synthetic test, too, but later it delivered superb performance in real-life tests. Perhaps the new Samsung is going to behave like the previous model?