I´m running kernel 2.6.8 from stable. After trying a lot of possible solutions, I faced (possibly) the solution. This machine has four memory slots and is able to deal with 4GB. When I use the four 1GB memory chips, I get the system running pretty slowly. If I take off one of the chips, everything turns out to be fine, and the raid performance goes up to 107MB/sec.
The kernel is compiled to support 4GB of memory. So why is this happening?
Thanks everyone,
Bruno.
On 11/1/05, Tom Vier <[EMAIL PROTECTED]> wrote:
On Tue, Nov 01, 2005 at 11:05:53AM -0200, Bruno Diniz wrote:
> /dev/sda:
> Timing buffered disk reads: 170 MB in 3.01 seconds = 56.41 MB/sec
> /dev/sdb:
> Timing buffered disk reads: 180 MB in 3.01 seconds = 59.81 MB/sec
> /dev/md0:
> Timing buffered disk reads: 134 MB in 3.04 seconds = 44.09 MB/sec
What kernel are you running? I don't have it that bad, with 2.6.13, but it's
definately slower than earlier 2.6 kernels. raid0 performance used to be
very good. raid1 has always sucked (reads no faster than a single drive).
/dev/md0:
Timing buffered disk reads: 322 MB in 3.01 seconds = 107.06 MB/sec
/dev/sda:
Timing buffered disk reads: 216 MB in 3.03 seconds = 71.37 MB/sec
/dev/sdb:
Timing buffered disk reads: 218 MB in 3.01 seconds = 72.53 MB/sec
--
Tom Vier < [EMAIL PROTECTED]>
DSA Key ID 0x15741ECE
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
--
Bruno Diniz de Paula
