Read also:

Introduction and summary
RAID0
RAID1

It is the most complicated mode for the controllers - there are 4 discs which form a Mirror array which, in its turn, consists of Stripe arrays of 2 discs. We will compare the test results with stripe arrays of 2 discs.

Ziff-Davis WinBench

Adaptec 1200A

AMI HyperDisk100

Promise FastTrak 100 TX2

All controllers try to maintain a constant linear reading speed. The Adaptec has an ideally even graph. But its speed is too low. The other two controllers have almost identical minimal reading speed on each part.

The CPU utilization didn't grow considerably. The access time has dropped, except the Adaptec's one.

In the FAT32 the overall performance has become worse in the High-End Disk WinMark test, and the Adaptec has suffered most of all. In the Business Disk WinMark the results are almost the same, it is only Adaptec which had worse scores. In the NTFS all controllers perform worse in both tests.

Intel IOMeter

The upper graph stands for a two-disc configuration, the lower one - for RAID0+1.

Server Pattern

Workstation Pattern

Database Pattern

The Promise has worsened its results, while the Adaptec and AMI perform better at large loads.

Streaming Read

Streaming Write

The results have risen at the expense of distribution of reading operations among two discs. The AMI is again ahead.

Random Read

Random Write

In the random write tests the AMI takes only the third place.

Conclusion

There is no an unquestionable leader. The AMI Hyperdisk100 and Promise FastTrak100 TX2 go on a par. Each has its own advantages. The Adaptec 1200A has shown an ideal graph and a quite low speed.

The complete results can be found in the tables.

Ziff-Davis WinBench 99
Intel IOMeter

• Server Pattern
• Workstation Pattern
• Database Pattern
• Streaming Read
• Streaming Write
• Random Read
• Random Write

Read also:

Introduction and summary
RAID1
RAID0+1