• 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.

Linux ICH8 benchmarks

D

drizzt81

[H]F Junkie
Joined
Jan 21, 2004
Messages
12,361
hi,

Since i just got done installing Linux on my new box (Suse 10.2 x64) I figured that it'd be nice to run some benches on my WD5000YS to see whether LSR is better than using the "firmware raid" provided by the ICH8r.I have manually made a 2.6.19 kernel and am running that now, since I thought that I needed that in order to use my ICH8r to its fullest potential. I have yet to try and get a firmware raid running. Since I'd rather not wait for a close to TiB sized array to initialize, I have decided to run my LSR tests in a 18GB RAID-5 array. I am mounting the disk(s) to /wd1

System:
C2D e6300 @ 2.8GHz, 2GB RAM, Gigabyte DS4.

Here are some bonnie++ numbers. I know that, like HDTach, this says little about real world performance, but at least it'll show how fast (or slow) my drives are:

Single WD5000YS (1) on ICH8r, in AHCI mode
Code: 
Version 1.01d       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
C2DLIN           4G 47181  21 46812  10 23043   2 64525  34 61850   4 248.0   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
C2DLIN,4G,47181,21,46812,10,23043,2,64525,34,61850,4,248.0,0,16,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++
Single WD5000YS (2) on ICH8r, in AHCI mode
Code: 
Version 1.01d       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
C2DLIN           4G 47057  21 47150  10 23097   3 71402  37 65461   5 247.8   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
C2DLIN,4G,47057,21,47150,10,23097,3,71402,37,65461,5,247.8,0,16,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++

Single WD5000YS (3) on ICH8r, in AHCI mode
Code: 
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
C2DLIN           4G 47057  21 47150  10 23097   3 71402  37 65461   5 247.8   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
C2DLIN,4G,47057,21,47150,10,23097,3,71402,37,65461,5,247.8,0,16,+++++,+++,+++++,+++,+++++,+++,+++++,+++


LSR RAID-5, 3 x WD5000YS on ICH8r, in AHCI mode, 18GB partition
Code: 
gwk@C2DLIN:~> bonnie++ -d /wd1
Version 1.01d       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
C2DLIN           4G 74239  33 77352  16 31929   5 92486  47 90164   9  80.0   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
C2DLIN,4G,74239,33,77352,16,31929,5,92486,47,90164,9,80.0,0,16,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++

What really baffles me about all these numbers is that it appears that my system can write faster than it reads. For a RAID-5 system, this makes ZERO sense to me. Anyone care to chime in?
 
A few notes:
Input is reading, Output is writing.

I recommend running the tests with a 10GB test file, that will help reduce RAM number inflation.
 
Dew said:
A few notes:
Input is reading, Output is writing.
Click to expand...
thanks, I was thinking in a disk-centric manner.

Here's my Seagate 7200.7 (160GB) for comparison. It is connected to the DS4's JMircon SATA controller via a SiI converter. I expected that a drive which is ~3 generations old was going to be easily beat by a recent gen drive.

Seagate 7200.7 160GB, JMicron SATA Controller
Code: 
Version 1.01d       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
C2DLIN           4G 31319  14 32023   7 11812   1 28375  14 35063   2 141.7   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
C2DLIN,4G,31319,14,32023,7,11812,1,28375,14,35063,2,141.7,0,16,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++
Single WD5000YS (1) on JMicron in AHCI mode
Code: 
Version 1.01d       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
C2DLIN           4G           46295  10 22740   3           63762   5 243.4   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
C2DLIN,4G,,,46295,10,22740,3,,,63762,5,243.4,0,16,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++
 
WD5000Ys on ICH8 (create)
Code: 
gwk@C2DLIN:/> dd if=/dev/zero of=/wd1/zerofile bs=1M count=8192
8192+0 records in
8192+0 records out
8589934592 bytes (8.6 GB) copied, 183.339 s, 46.9 MB/s

WD5000YS on JMicron (create)
Code: 
gwk@C2DLIN:/test> dd if=/dev/zero of=/wd1/zerofile bs=1M count=8192
8192+0 records in
8192+0 records out
8589934592 bytes (8.6 GB) copied, 183.963 s, 46.7 MB/s
WD5000YS on ICH8 (read)
Code: 
gwk@C2DLIN:/> dd if=/wd1/zerofile of=/dev/null bs=128k
65536+0 records in
65536+0 records out
8589934592 bytes (8.6 GB) copied, 138.456 s, 62.0 MB/s
WD5000YS on JMicron (read)
Code: 
gwk@C2DLIN:/> dd if=/wd1/zerofile of=/dev/null bs=128k
65536+0 records in
65536+0 records out
8589934592 bytes (8.6 GB) copied, 131.002 s, 65.6 MB/s
7200.10 on JMicron (create)
Code: 
gwk@C2DLIN:/test> dd if=/dev/zero of=./zerofile bs=1M count=8192
8192+0 records in
8192+0 records out
8589934592 bytes (8.6 GB) copied, 253.718 s, 33.9 MB/s
7200.10 on JMicron (read)
Code: 
gwk@C2DLIN:/test> dd of=/dev/null if=./zerofile bs=[color=red]1M(whoops)[/color] count=8192
8192+0 records in
8192+0 records out
8589934592 bytes (8.6 GB) copied, 226.665 s, 37.9 MB/s
7200.10 on JMicron (read)
Code: 
gwk@C2DLIN:/test> dd of=/dev/null if=./zerofile bs=128k
65536+0 records in
65536+0 records out
8589934592 bytes (8.6 GB) copied, 242.422 s, 35.4 MB/s
 
You must log in or register to reply here.
Back
Top