• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Slow RAID 0 performance, I need some help.

Megatomic

Megatomic

Lifer
Epox 8K3A+ with HPT372 ver. 2.32 with 2.32 drivers
2 x Maxtor 40GB 7200RPM ATA133 HDDs in RAID 0
1 x 512MB Mushkin CAS2 PC2100 (fast timings)
WinXP Pro

When I ran SiSoft Sandra 2002 on a single Maxtor 40GB 7200RPM ATA133 drive I scored about 26,500KB/sec. I then added the second drive, setup the RAID array, and reinstalled WinXP Pro. Used the latest Highpoint drivers and installed the 4.43 Via 4-in-1 patch. When I installed SiSoft Sandra 2002 again and reran the Filesystem benchmark on the RAID array I got the same exact score. Isn't RAID 0 supposed to be faster?

I then made sure that optimal performance was enabled and it was (it was actually grayed out so I couldn't adjust it if I wanted to) and I made the RAID array a Dynamic disk. I also defragged the drive with Diskeeper 7 Workstation. When I ran the benchmark again I had the same score.

What do I need to do to get this thing running as fast as it should be able to run? I will value any constructive comment you have to offer. 🙂 Thanks.
 
I read on THIS PAGE that best performance is achieved if you use one drive on it's own channel and use both channels to make the RAID 0 array. Kinda makes sense. But the way I did it was to put the two drives I made the array out of on the same channel and created the array then. I still think I should see a noticeable increase in throughput even if my array is not in the most optimized configuration.

What do you all think?


Edit: Fixed link.
 
Well, the cables I am currently using are about the best you can expect for ATA133 ribbon cables, but you also recommend separating the drives onto separate channels. If I do that, will I have to rebuild the array and reinstall windows again?

And for benching the array, you are recommending HDTach... Is that the standard used?
 
I'm not sure if you have to rebuild the array, I suppose it depends on the controller. I'm afraid I can't say for sure. 🙁

I've never trusted SiSoft Sandra, ppl around here tend to use HDTach when it comes to HDs.
 
I downloaded HDTach but I couldn't do anything with it. I need to buy a license or some such first. 🙁 Is it worth the money?

And thanks for the tip on the driver WobbleWobble, I'll give it a shot. Couldn't hurt at this point.
 
I actually had a similar problem with my RAID-0 setup.

I have an Asus A7V133 with a Promise RAID controller and 2 Maxtor D740X 40GB HDDs (same as yours).

When I first benchmarked, I also noticed relatively poor performance. That Via RAID driver really helped a lot by about 50% for me.
 
I read on THIS PAGE that best performance is achieved if you use one drive on it's own channel and use both channels to make the RAID 0 array. Kinda makes sense. But the way I did it was to put the two drives I made the array out of on the same channel and created the array then. I still think I should see a noticeable increase in throughput even if my array is not in the most optimized configuration.
Click to expand...

When you have two devices on the same channel only one can be accessed at a time. This is fine for two hard drives in a non-raid array but the very point of raid 0 is being able to access both drives at the same time. Since you can't do this with both drives on the same channel performance is poor at best. You need two seperate channels.
 
Originally posted by: VTrider
Originally posted by: WobbleWobble
Try the RAID Performance Patch from Via.

Link to the RAID Performance Patch
Click to expand...

According to that page, this patch doesn't work with the newer VT8235 Southbridge, I'm pretty sure that Epox 8K3A+ has that chipset? Correct?



-VTrider
Click to expand...

I actually have the VT8233A southbridge chip. The patch is fine for my board but any newer Via based board is excluded. I did apply the patch and it made a slight difference in HD Tach but nothing I see in real world performance. I still appreciate the linkage there WobbleWobble.

ams30gts - all the Abit KR7A boards have southbridge chips that are older than the VT8235, you'll be fine to try the patch.

 
You must log in or register to reply here.

TRENDING THREADS

Back
Top