viz Guest
|
Posted: Fri Nov 14, 2008 6:31 pm Post subject: Re: Intel Quad Core performance on CentOS 5.2 is slower |
|
|
On Nov 13, 1:08 am, habibielwa7id <fouad...@gmail.com> wrote:
| Quote: | The speed of the hard disk should be better than this really, The
problem you have happened to me before with HP system and fedora core
6, In the beginning really I thought that my hard disk isn't working
fine and thought to replace it, And when I changed it this didn't help
anyway, It was a driver issue, When I updated to fedora 9 it went
really fine and working fast and fine tell now, The hard disk worked
properly, Try some commands I think they will help you to have more
information about the issue,
dmesg |grep sda for example
lspci -v
As I can remember the driver issue wasn't related to the hard disk it
self, It was related to the board IDE subsystem,
You may try smartmontools, It should give you more information about
the status of the hard disk.
Regards,
|
Thanks habibielwa7id. Even though I witnessed the HD performance going
up 25x, I don't know if it can be bettered further. The datasheet for
the hard disk promises a performance of 78 MB/s max (= "maximum
sustained data transfer rate" property in the datasheet), while hdparm
shows the performance around 75 MB/s. Moreover, I got these figures
while retaining my original kernel 2.6.18. Would the performance be
even better if I upgrade the kernel (though I am satisfied with the
current figures)?
Here are the outputs of the commands you suggested (for lspci, see my
previous post):
$ dmesg |grep sda
SCSI device sda: 625140335 512-byte hdwr sectors (320072 MB)
sda: Write Protect is off
sda: Mode Sense: 00 3a 00 00
SCSI device sda: drive cache: write back
SCSI device sda: 625140335 512-byte hdwr sectors (320072 MB)
sda: Write Protect is off
sda: Mode Sense: 00 3a 00 00
SCSI device sda: drive cache: write back
sda: sda1 sda2 sda3 sda4 < sda5 >
sd 0:0:0:0: Attached scsi disk sda
EXT3 FS on sda2, internal journal
EXT3 FS on sda5, internal journal
EXT3 FS on sda1, internal journal
Adding 4192956k swap on /dev/sda3. Priority:-1 extents:1 across:
4192956k
Sincerely,
Kumar Vijay Mishra. |
|
habibielwa7id Guest
|
Posted: Sun Nov 16, 2008 9:16 am Post subject: Re: Intel Quad Core performance on CentOS 5.2 is slower |
|
|
On Nov 14, 8:31 pm, viz <vizz...@gmail.com> wrote:
| Quote: | On Nov 13, 1:08 am, habibielwa7id <fouad...@gmail.com> wrote:
The speed of the hard disk should be better than this really, The
problem you have happened to me before with HP system and fedora core
6, In the beginning really I thought that my hard disk isn't working
fine and thought to replace it, And when I changed it this didn't help
anyway, It was a driver issue, When I updated to fedora 9 it went
really fine and working fast and fine tell now, The hard disk worked
properly, Try some commands I think they will help you to have more
information about the issue,
dmesg |grep sda for example
lspci -v
As I can remember the driver issue wasn't related to the hard disk it
self, It was related to the board IDE subsystem,
You may try smartmontools, It should give you more information about
the status of the hard disk.
Regards,
Thanks habibielwa7id. Even though I witnessed the HD performance going
up 25x, I don't know if it can be bettered further. The datasheet for
the hard disk promises a performance of 78 MB/s max (= "maximum
sustained data transfer rate" property in the datasheet), while hdparm
shows the performance around 75 MB/s. Moreover, I got these figures
while retaining my original kernel 2.6.18. Would the performance be
even better if I upgrade the kernel (though I am satisfied with the
current figures)?
Here are the outputs of the commands you suggested (for lspci, see my
previous post):
$ dmesg |grep sda
SCSI device sda: 625140335 512-byte hdwr sectors (320072 MB)
sda: Write Protect is off
sda: Mode Sense: 00 3a 00 00
SCSI device sda: drive cache: write back
SCSI device sda: 625140335 512-byte hdwr sectors (320072 MB)
sda: Write Protect is off
sda: Mode Sense: 00 3a 00 00
SCSI device sda: drive cache: write back
sda: sda1 sda2 sda3 sda4 < sda5
sd 0:0:0:0: Attached scsi disk sda
EXT3 FS on sda2, internal journal
EXT3 FS on sda5, internal journal
EXT3 FS on sda1, internal journal
Adding 4192956k swap on /dev/sda3. Priority:-1 extents:1 across:
4192956k
Sincerely,
Kumar Vijay Mishra.
|
I think the speed you could obtained later is good, This is a good
speed as I saw many systems around this speed, May I say some thing
here,
I really see that the speed of the not original computers is better
in general than the original computers, This doesn't mean that the not
original computers are better than the original computers, I think the
original computers more stable and safer And can carry more hard work,
But the only advantage of the not original computers as I saw many
times is the speed,
for example I saw a speed of 100 MB with sheep and not original pcs,
hahah Although I work on expensive original servers but I havn't seen
this speed with these computers,
By the way, Don't measure the speed of your pc while it's under heavy
load, It will be worse of course if you compare it with the output of
the hdparm utility when your pc isn't under a heavy load.
Regards, |
|