Unsolved
This post is more than 5 years old
5 Posts
0
110698
Perc 4E/DC
Dell Perc 4e/dc flashed Intel SRCU42E firmware and bios.
Oops! Not working!Messege alert!
RAID adapter
Raid Key OEM Authentication Error!!!
Install the correct RAID key for this System.
If problems persist contact Tech Support.
Press Any Key to Continue...
-
How to turn Dell bios?
Oops! Not working!Messege alert!
RAID adapter
Raid Key OEM Authentication Error!!!
Install the correct RAID key for this System.
If problems persist contact Tech Support.
Press Any Key to Continue...
-
How to turn Dell bios?
Please Need help!!
jbmaciel
6 Posts
0
July 23rd, 2008 18:00
I did test with HDtach and as I remember the transfer speed was somenthing between 40-50 MB/s. The Sandra reported ~70 MB/s.
The HDs are SCA (Fujitsu MAX3147NC), so there´s no jumpers. The jumpers are in the adapter (XPL-065B ver. 1.1) which I think is correct.
Regards,
Joselito Maciel.
Ian Mapleson
22 Posts
0
July 24th, 2008 09:00
I'm at a loss to now what to suggest without being able to inspect the setup personally.
Strange part is, if the SCSI bus was definitely running at 40, then the disks shouldn't
be able to go any faster than about 35 to 38, unless buffering/cache is bursting the
speed beyond that, in which case the tests are not directly testing the drives.
No luck finding a Tyan user forum?
Ian.
jbmaciel
6 Posts
0
July 24th, 2008 10:00
Ok, I forgot to tell you. The two Hds are running on RAID 0 mode. That's why the data transfer was higher than 40MB/s.
I just solve the problem. I've got two new adapters from a local store. The salesman told me that these were U320 complaint, so I grabbed two to test. They work, now the card says 320MB/s.
The Syssoft Sandra is reporting a average of 240MB/s. Much better.
Regards,
Joselito Maciel.
Ian Mapleson
22 Posts
0
July 24th, 2008 11:00
Excellent! 8) Glad to hear that. Indeed, sounds like the old adapters were like the type
I mentioned, not rated beyond 40. Hard to tell which ones are and which aren't. I guess
if they're cheap, then beware.
I'm in the process of setting up my research RAID, 4 x Sagitta units, 48 disks total. Hoping
to get past 600MB/sec on a 10-year-old SGI, should be a laugh. :D
Found a cool extension lead that has a radio remote control for the 4 sockets. Very handy!
Means I don't have to reach behind the units to turn them on/off. Only $30 aswell, was a
half-price offer.
Ian.
jbmaciel
6 Posts
0
July 24th, 2008 14:00
600MB/s? Wonderfull.
I have in my machine 4x WD 250GB 3Gb/s. I mean 1TB in RAID 0, over standard PCI. The throughput is around 120 MB/s (PCI bus is the bottleneck here).
The U320 SCSIs are in the same machine (actually the the sustained read is about 140 MB/s, not 240), over PCIe 4X. If I remember, 240 MB/s was the burst read.
They are 15K 146GB (MAX3147NC), from Fujitsu. So, they are quiter old. The 140 MB/s are expected, but the burst should be higher I think.
Regards,
Joselito Maciel.
Ian Mapleson
22 Posts
0
July 24th, 2008 23:00
> 600MB/s? Wonderfull.
I'm trying to find out just how fast the I/O can be on the system. I
know the memory bus does not support more than about 800MB/sec
sustained (remember this a system from 1996 :D) so if I can get
600MB/sec then I'll be pleased. I use lots of old 73GB 1.6" disks
for testing (because they're cheap), though I could get the same
results if I used a small number of modern drives, but newer drives
cost too much.
Funny thing, results can be better with 10K drives sometimes. Admin
guy at SUN reckons this can happen when read requests don't quite
match the coincidental timing of where the heads are positioned at
the time of the request, ie. 15K drives may effectively be spinning
too fast for maximum efficiency in certain situations. Varies though.
> I have in my machine 4x WD 250GB 3Gb/s. I mean 1TB in RAID 0, over
> standard PCI. The throughput is around 120 MB/s (PCI bus is the
> bottleneck here).
Yeah, I had the same problem with my old mbd, LSI PCIX card limited
by PCI bus. Once moved to a proper PCIX slot, the test results more
than tripled.
> They are 15K 146GB (MAX3147NC), from Fujitsu. So, they are quiter
> old. ...
I have two similar drives, the MAU version (SUN-badged MAU3147NC).
Did have three, but one died. I've been switching to Maxtors though
as they're quite a bit faster (Maxtor Atlas 15K II, 8K147J0).
> ... The 140 MB/s are expected, but the burst should be higher I
> think.
Perhaps limited by cache/RAM issues? Hard to say. The newer RAM
on the LSI PCIe card I have did give much higher cached I/O than
the LSI PCIX version.
Ian.