Database Patterns

In the Database pattern the drive is processing a stream of requests to read and write 8KB random-address data blocks. The ratio of read to write requests is changing from 0% to 100% with a step of 10% throughout the test while the request queue depth varies from 1 to 256.

You can click the following links to view the tabled results:

Intel IOMeter: Database (details)

We will build diagrams for request queue depths of 1, 16 and 256. There are too many results, so we divide them into two diagrams here and in some of the later tests.

At a queue depth of 1 request, the Seagate proves to be the fastest in pure reading but accelerates less rapidly than the other HDDs as the percentage of write requests is growing up. The 320GB model from Western Digital goes ahead at 30% write requests and is joined by the 250GB drive from the same brand at 40% and more writes. When there are only write requests to be processed, the HDDs from Western Digital are accompanied with the HDDs from Fujitsu and Hitachi which are very effective at high percentages of writes. The 250GB Toshiba and Samsung are the losers of this test.

When the queue is 16 requests long, the request reordering algorithms begin to affect the drives’ performance. As a result, the Seagate loses its first place, giving way to the new 320GB models from Toshiba and Western Digital. The latter is the absolute winner at any combination of load. The drives from Hitachi and Fujitsu still feel good at high percentages of writes, but while the Hitachi is average at other loads, the Fujitsu is downright slow at mixed loads. The 250GB Toshiba is the worst one under every load.

The 320GB Western Digital enjoys a huge lead over the others at very long queue depths. It is only at 100% writes that it is overtaken by the Hitachi. The new 52 series Toshiba feels good at very high percentages of reads but is not brilliant at all at writing. The 250GB Toshiba, Samsung and, perhaps, Fujitsu don’t perform well here. The Fujitsu is good at writing but very poor at reading.