Navigate: Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors Home PC Hardware from Made-in-China.com Search: Enter your search terms   Web www.lostcircuits.com Submit search form	.
Prices:

LOSTCIRCUITS		SHORTCUTS: The Challenge Fujitsus and Promises Assembling the RAID Building the System Test configuration HDTach vs. WB99 ATTO vs. Sandra Back to HDTach, WB99 Network File Copy Conclusions Fujitsu Drives Comments?
Internal 4-Way SATA RAID For The XPC     A Concept Study
(Review by MS, April 20, 2003)

If it is the overall media transfer rate that causes problems with HDTach, we should be able to get more reliable numbers and smoother graphs by moving to a 2 disk RAID 0 or a single drive. Showing a single drive benchmark here.. oh please, the purpose was to show a RAID setup... suffice it to say that the OD sequential performance is about 24 MB/s. But then there is a four disk RAID 01 (2+2 in Promise nomenclature) setup which is actually what I would personally run if this were my day-to day system. In other words, it is a RAID 01 (striped across two drives and mirrored on the other two drives), how much swell drive goodness can you have in one box?

Looks like our assumption was correct, the OD performance is at 47 MB/sec exactly where it is expected, the Random accesses are a bit sluggish, which is normal with 4 drives around and the burst rate... for some reason, the patch did not work so we can't really tell. Bottom line is that it looks like the drives and the interface card are working as advertised, if not better but that some benchmarks have aged and do no longer give accurate readings. In addition, the minimum and average performance are completely meaningless.

Back to the 4 disk RAID 0

Winbench99 2.0

Nothing replaces Winbench99 disk playback as drive benchmark as long as it is run on a FAT file system which is where it was recorded on originally. One thing to keep in mind here is that the business performance in a well-greased system is almost directly proportional to the size of the disk cache. It sounds a bit oversimplified but it is a good rule of thumb. Since the four drives have a combined cache size of 2 MB, a Business Disk Winmark score of a bit over 10,000 is right on the money.

Winbench99 2.0 disk Winmark scores in a 4 disk RAID 0 configuration. The Frontpage98 score was adjusted to 10% of its actual value to match the remaining scores for better resolution of the graph.

next page:    => File Copy -> Conclusion =>