pients of list ORACLE-L <[EMAIL PROTECTED]>
cc:
Subject:RE: SunOS 5.8 I/O buffer size?
Tim, I don't believe that you can see that by strace. The whole
thing is happening entirely in the kernel. The problem is that
each device has a fixed size buffer alloca
June 13, 2002 10:13 PM
> To: Multiple recipients of list ORACLE-L
> Subject: Re: SunOS 5.8 I/O buffer size?
>
>
> Carmen et al,
>
> Did some testing on this. On solaris 2.8 there doesn't seem to be any
> upper-limit on single, atomic disk I/O, probably because
>
Title: RE: SunOS 5.8 I/O buffer size?
Yeah, I've been using this device for backups for
years. Always get very good performance.
(This is a joke folks!)
Matt Adams - GE Appliances - [EMAIL PROTECTED]
When someone says "I want a programming language in which I
only need
Hey Tim!
That /dev/null device is really fast! Solaris wrote 10 megabytes
to it in 0.5 seconds.
Where can I get one of those?
Jared
;)
On Thursday 13 June 2002 19:13, Tim Gorman wrote:
> Carmen et al,
>
> Did some testing on this. On solaris 2.8 there doesn't seem to be any
> upper-l
Live and learn! I'd never heard of SSTIOMAX, but there's a decent MetaLink
article on it (#131530.1)...
I tried using "nm" and "strings" on the "oracle" executable and many of the
".a" and ".so" objects and couldn't find it mentioned, so it must be a
"#define" compiler directive in the source co
On 2.7
0.8128 read(3, "\0\0\0\0\0\002\0\0\0F0\0"..,
10485760) = 10485760
0.1003 write(4, "\0\0\0\0\0\002\0\0\0F0\0"..,
10485760) = 10485760
0.7603 read(3, "\0\00386\0\0 P\01BAFD0FA"..,
10485760) = 10485760
0.1039 write(4, "\0\00386\0\0 P\01BAFD0FA"..,
10485760) = 10485760
1.0187 read(3, "\0\
Hi Tim
Here are numbers for a Redhat 7.2 Linux with kernel 2.4.17 using
LVM1.0.3 on the disk.
There is something rotten here since the read time is not increasing
with a factor 10 as I would
expect the write time is behaving close to expected.
If you want the full strace drop a mail.
1 Mb
Carmen et al,
Did some testing on this. On solaris 2.8 there doesn't seem to be any
upper-limit on single, atomic disk I/O, probably because someone finally got
smart and started allocating memory buffers for I/O dynamically using the
"malloc()" package instead of hard-coding a fixed-length vari
3 PM
> To: Multiple recipients of list ORACLE-L
> Subject: Re: SunOS 5.8 I/O buffer size?
>
>
> Amazing how so many can't tell a Ferrari from a Porsche.
>
> Not you Mladen, I realize that's why "Ferrari" was in quotes.
>
> Jared
>
>
ecipients of list ORACLE-L <[EMAIL PROTECTED]>
cc:
Subject: Re: SunOS 5.8 I/O buffer size?
The "IO buffer size" is the maximal size of a single, atomic disk IO,
usually
either 128k or 256k. All IO requests larger then that will be
internally broken
into m
The "IO buffer size" is the maximal size of a single, atomic disk IO,
usually
either 128k or 256k. All IO requests larger then that will be
internally broken
into multiple requests. As my Solaris is getting rusty (working with
HP-UX and
Linux now) I cannot remember the exact parameter name. It
11 matches
Mail list logo
