• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

37.4MB/s RAID-5?

TeeJayHoward

TeeJayHoward

Limpness Supreme
2FA
Joined
Feb 8, 2005
Messages
13,351
hdtach1wk.png

Surely this can't be right... Can it?
 
drizzt81 said:
it depends
Click to expand...
I'm going to assume it depends on a variety of factors, so I'll expand.
8x320GB RAID-5 off a HighPoint 2220 in a 32-bit PCI slot on a BFG BFGRNF4U motherboard that's slowly dying, using an Opteron 165 at stock speeds. The array has a single folder on it at the moment with 2.58GB of data inside it. It was initialized via the HighPoint RAID Management software inside of Windows XP x64. At the time of the test, the only other running programs were Windows Messenger and uTorrent, which was writing to a seperate SATA drive on the motherboard's chipset at approximately 200KB/s. There are no other PCI devices installed.
 
TeeJayHoward said:
I'm going to assume it depends on a variety of factors, so I'll expand.
8x320GB RAID-5 off a HighPoint 2220 in a 32-bit PCI slot on a BFG BFGRNF4U motherboard that's slowly dying, using an Opteron 165 at stock speeds. The array has a single folder on it at the moment with 2.58GB of data inside it. It was initialized via the HighPoint RAID Management software inside of Windows XP x64. At the time of the test, the only other running programs were Windows Messenger and uTorrent, which was writing to a seperate SATA drive on the motherboard's chipset at approximately 200KB/s. There are no other PCI devices installed.
Click to expand...
Considering that you have a relatively new RAID card and nothing on the PCI bus: yes that score is really low.
Here are some questions:
Is the array initialization complete? (I am sure you have, but asking did not hurt anyone)
How about the CPU utilization?
Did you re-run the test without anything else running?
 
drizzt81 said:
Considering that you have a relatively new RAID card and nothing on the PCI bus: yes that score is really low.
Here are some questions:
Is the array initialization complete? (I am sure you have, but asking did not hurt anyone)
How about the CPU utilization?
Did you re-run the test without anything else running?
Click to expand...

The initialization completed yesterday, yes. Something I should note: Due to my mobo being finicky, I have to disable the RAID array on the card's bios to boot, and then re-enable it inside windows. Could this be causing the problem?

As for CPU utilization, during this test, the CPU did practically nothing.

I'm re-running the test now.
 
I ran a 32MB test instead of the 8MB test. Here are the results:
raid5ju.png

I'm only more confused. On a side note - HTTach is currently maxing out one core on my processor. It wasn't during the test, but while displaying the results, I'm at 50% usage. WTF?

Here's a pic of the two compared side by side (I ran the 8MB and 32MB tests again):
raid4br.png
 
I am going out on a limb here, but if your mobo is dieing, it may be the cause of all of this.

note: I am no expert, maybe this is a well known problem that someone with more specific knowledge will easily troubleshoot for you.
 
That's *really* weird. Compare the first and second runs of the 32 MB test; they're at completely different heights (average 75 max 90 versus 52/60). HDTach is usually pretty good about being consistent, I don't know why it would act so strangely.

 
Since those are not RE drives, you may have on that is sending an extended error time out that is not quite long enough to fail the drive out of the array, then you might not ever see the problem.

This is why WD makes RE drives.. their non RE drives do not work well in RAID arrays.

==>Lazn
 
Yay. More good news. I just discovered that any data saved on the array gets corrupted. Ugh. I cannot WAIT until my replacement board arrives.
 
TeeJayHoward said:
Yay. More good news. I just discovered that any data saved on the array gets corrupted. Ugh. I cannot WAIT until my replacement board arrives.
Click to expand...
New board arrived. DFI LanParty Ultra-D.
hdtach3xy.png


Hmm. Maybe it's not the board's fault, but rather the controller's? Should I contact ZZF and see about a replacement? Are there any tests I should run? The read speeds are about on par with a single drive and the burst's slightly lower... Somehow I thought they'd be higher.
 
That's really odd. Can anyone else with one of these under Windows post their setup and HDTach?

Maybe running IOMeter? Sequential 64k reads will probably yield the same thing, but it's worth checking...

 
You must log in or register to reply here.
Back
Top