Lukas Ertl [EMAIL PROTECTED] writes:
Anyone has an idea what's going wrong here? (Apart from me doing bullsh*t
benchmarking :-) .)
Just out of curiosity, try again with prime stripe sizes (31, 61, 127,
257, 509) or at least odd ones (31, 63, 127, 255, 511).
DES
--
Dag-Erling Smørgrav - [EMAIL
Lukas Ertl [EMAIL PROTECTED] writes:
I'm currently testing with prime stripe sizes, but it doesn't seem to
help. I additionally added options AHC_ALLOW_MEMIO to the kernel, and it
has raised write performance in the single-disk case (although I'm not
happy with that one either; I expected a
On Sun, 30 Mar 2003, Dag-Erling Smørgrav wrote:
Lukas Ertl [EMAIL PROTECTED] writes:
I'm currently testing with prime stripe sizes, but it doesn't seem to
help. I additionally added options AHC_ALLOW_MEMIO to the kernel, and it
has raised write performance in the single-disk case (although
Lukas Ertl [EMAIL PROTECTED] writes:
On Sun, 30 Mar 2003, Dag-Erling Smørgrav wrote:
Does the data sheet for your disk indicate that it can in fact write
much faster than that?
Well, the HP/Compaq webpages are full of marketing speech wrt that, but
since these disks are U320 disks, they
On Sunday, 30 March 2003 at 14:08:24 +0200, Dag-Erling Smrgrav wrote:
Lukas Ertl [EMAIL PROTECTED] writes:
Anyone has an idea what's going wrong here? (Apart from me doing bullsh*t
benchmarking :-) .)
Just out of curiosity, try again with prime stripe sizes (31, 61, 127,
257, 509) or at
On Sunday, 30 March 2003 at 16:30:17 +0200, Dag-Erling Smrgrav wrote:
Anti [EMAIL PROTECTED] writes:
try benchmarking ccd as well to see how it compares? and try some big stripe
sizes (4093, 8191, 12281)?
I believe Greg showed in his USENIX paper on Vinum that large stripe
sizes aren't
On Sunday, 30 March 2003 at 18:10:07 +0200, Dag-Erling Smrgrav wrote:
Lukas Ertl [EMAIL PROTECTED] writes:
I'm currently testing with prime stripe sizes, but it doesn't seem to
help. I additionally added options AHC_ALLOW_MEMIO to the kernel, and it
has raised write performance in the
* skip the following three paragraphs if you're not interested in the
politics *
This is funny. Not funny "ha ha", but I think it's actually indicative of
the high quality of free software: we have to argue whether pre-release
features/functionality should be used in critical production
At 02:09 PM 9/15/99 -0700, Mike Smith wrote:
Perhaps we're seeing this differently, but "sworn off vinum" says to me
"never using it again", and that's the point I was trying to make.
People try something that's still under development, get bitten, and
then claim forever after that the
At 7:11 PM -0500 1999/9/15, Jeffrey J. Mountin wrote:
Well Greg considers it a "release" anyway and no longer alpha. Surely it
may be "under development" still, but then isn't -stable?
My apologies. I was under the impression that it was still
considered alpha code, but then I
At 03:45 AM 9/16/99 +0800, Michael Robinson wrote:
Also, keep in mind that vinum does very little in the way of sanity checking.
It's very happy to trash your data if you tell it to do something even a
little bit unusual.
Like anything it will only do what you tell it, so in a way I'm of
As I recall, Adrian Filipi-Martin wrote:
This is in line with what I would expect for RAID 0 and 1. CPU
usage just makes the I/O performance measurements more compelling. CPU
will also be more interesting when it comes to measuring the RAID 5
performance later.
And you'd
"Jeffrey J. Mountin" [EMAIL PROTECTED] writes:
Like anything it will only do what you tell it, so in a way I'm of agreement.
My biggest objection to vinum is that it will do what you tell it to,
and then complain afterwards, rather than complain about it at the time
you try to do it. Two good
Mike Smith [EMAIL PROTECTED] writes:
Also, keep in mind that vinum does very little in the way of sanity checking.
It's very happy to trash your data if you tell it to do something even a
little bit unusual.
After getting seriously burned (fortunately, during initial system
configuration),
At 02:30 AM 9/16/99 +0200, Brad Knowles wrote:
My apologies. I was under the impression that it was still
considered alpha code, but then I guess if it was it probably
wouldn't have migrated to the -STABLE tree, now would it?
None needed, but we're not the only ones that assumed that
15 matches
Mail list logo
