?
Solved

Adaptec Raid on Linux slow

Posted on 2003-03-10
7
Medium Priority
?
504 Views
Last Modified: 2006-11-17
Hi all,

We installed an Adaptec 2200s Ultra320 controller,
with two Seagate Ultra320 HDs, 15000 rpm, 73 GB.
Configured as Hardware-raid 1.
OS is RedHat 7.3.
Installed the drivers that came with the controller.

We ran hdparm -tT against it and it said:
buffer-cache reads about 110 MB/sec
buffered disk reads about 30 MB/sec.
which seemed a little slow to me.
(Hardware IDE Raid 1 in other server
makes 400 MB/sec and 45 MB/sec.)

So we ran tiobench against it.
We're not sure we understand what we see.
Could someone comment on the results?
We ran tiobench without any parameters,
so they were automatically addedas follows:
-t 1 -f 1278 -r 4000 -b 4096 -d . -T

Thanks in advance!

Here's the result:


Sequential Reads
File Blk Num Avg Maximum Lat% Lat% CPU
Size Size Thr Rate (CPU%) Latency Latency >2s >10s Eff
------ ----- --- ------ ------ --------- ----------- -------- -------- -----
1278 4096 1 30.21 17.25% 0.128 64.00 0.00000 0.00000 175
1278 4096 2 41.65 24.86% 0.186 150.05 0.00000 0.00000 167
1278 4096 4 41.01 23.84% 0.371 268.59 0.00000 0.00000 172
1278 4096 8 39.52 27.59% 0.717 622.73 0.00000 0.00000 143

Random Reads
File Blk Num Avg Maximum Lat% Lat% CPU
Size Size Thr Rate (CPU%) Latency Latency >2s >10s Eff
------ ----- --- ------ ------ --------- ----------- -------- -------- -----
1278 4096 1 1.51 2.023% 2.591 23.71 0.00000 0.00000 74
1278 4096 2 1.77 1.356% 4.407 32.88 0.00000 0.00000 130
1278 4096 4 2.39 2.597% 6.303 61.02 0.00000 0.00000 92
1278 4096 8 2.33 2.088% 12.287 218.09 0.00000 0.00000 112

Sequential Writes
File Blk Num Avg Maximum Lat% Lat% CPU
Size Size Thr Rate (CPU%) Latency Latency >2s >10s Eff
------ ----- --- ------ ------ --------- ----------- -------- -------- -----
1278 4096 1 11.43 20.96% 0.249 3420.60 0.00183 0.00000 55
1278 4096 2 14.25 25.85% 0.418 3229.35 0.00245 0.00000 55
1278 4096 4 11.35 20.02% 0.956 6423.05 0.00980 0.00000 57
1278 4096 8 14.17 25.14% 1.555 6771.49 0.02058 0.00000 56

Random Writes
File Blk Num Avg Maximum Lat% Lat% CPU
Size Size Thr Rate (CPU%) Latency Latency >2s >10s Eff
------ ----- --- ------ ------ --------- ----------- -------- -------- -----
1278 4096 1 1.95 4.377% 0.043 3.95 0.00000 0.00000 45
1278 4096 2 2.34 5.086% 0.042 4.11 0.00000 0.00000 46
1278 4096 4 2.02 4.387% 0.042 4.09 0.00000 0.00000 46
1278 4096 8 2.38 5.323% 0.041 0.17 0.00000 0.00000 45
0
Comment
Question by:Oli2
[X]
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
7 Comments
 
LVL 1

Author Comment

by:Oli2
ID: 8101912
If you copy+paste the results to a standard editor, it becomes much more readable...  ;-)

regards,
Oli
0
 
LVL 10

Expert Comment

by:kiranghag
ID: 8103107
i had come accross one question here that concluded that VIA chipset are not good for RAID as they tend to limit transfer rates in raid volume..
do you have via chipset????
0
 
LVL 4

Expert Comment

by:1175089
ID: 8105069
This is not a real true, the problem was in VIA's implementation in PCI bus mastering, so if you have integrated (or external, they use the same PCI connect bus) RAID controller, data transfer was lower, than other chipset makers. In these samples you use data chunks with size of 4096 bytes, are the same in the other server? Other cause may be difference in size/type of RAID controller memory. (Write through instead of write back cache i.e, bad (old) RAID drivers i.e
0
Industry Leaders: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

 
LVL 1

Author Comment

by:Oli2
ID: 8239684
Thanks for the help guys, it turned out the problem was the wrong kind of SCSI-terminator.

regards,
Oli
0
 
LVL 6

Expert Comment

by:akboss
ID: 10292412

============================
No comment has been added lately, so it's time to clean up this TA.
I will leave a recommendation in the Cleanup topic area that this question is:
PAQ/Refund Points
Please leave any comments here within the next seven days.
 
PLEASE DO NOT ACCEPT THIS COMMENT AS AN ANSWER!
 
akboss
EE Cleanup Volunteer
============================
0
 

Accepted Solution

by:
amp072397 earned 0 total points
ID: 10357856
PAQed, with points refunded (275)

amp
Community Support Cleanup Moderator
0

Featured Post

What does it mean to be "Always On"?

Is your cloud always on? With an Always On cloud you won't have to worry about downtime for maintenance or software application code updates, ensuring that your bottom line isn't affected.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Monitor input from a computer is usually nothing special.  In this instance it prevented anyone from using the computer.  This was a preconfiguration that didn't work.
This paper addresses the security of Sennheiser DECT Contact Center and Office (CC&O) headsets. It describes the DECT security chain comprised of “Pairing”, “Per Call Authentication” and “Encryption”, which are all part of the standard DECT protocol.
In this video you will find out how to export Office 365 mailboxes using the built in eDiscovery tool. Bear in mind that although this method might be useful in some cases, using PST files as Office 365 backup is troublesome in a long run (more on t…
Have you created a query with information for a calendar? ... and then, abra-cadabra, the calendar is done?! I am going to show you how to make that happen. Visualize your data!  ... really see it To use the code to create a calendar from a q…
Suggested Courses

762 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question