I have RH 7.2 as a guest under VM 2.4 which is in turn running in an LPAR
on a MP3000-P30 so I thought I'd do some tests.  The Linux guest uses
internal disk for storage and has 32MB of RAM and 128MB of V-Disk swap
space.  The VM system is running with a very low 'weight' (dead last
priority) and it's the middle of the day but it's a lightly loaded system
most of the time.  Anyway, take these readings with a shaker of salt ;-)

Using Bonnie++ 1.02a

[zphelj@tux1 bontest]$ uname -a
Linux tux1 2.4.9-17 #1 SMP Fri Nov 23 19:45:36 CET 2001 s390 unknown
[zphelj@tux1 bontest]$ /usr/local/sbin/bonnie++ -s 200
Writing with putc()...done
Writing intelligently...done
Rewriting...done
Reading with getc()...done
Reading intelligently...done
start 'em...done...done...done...
Create files in sequential order...done.
Stat files in sequential order...done.
Delete files in sequential order...done.
Create files in random order...done.
Stat files in random order...done.
Delete files in random order...done.
Version 1.02a       ------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
tux1           200M  1040  96  3938  13  3244   7   941  86  7269  11 231.1
8
                    ------Sequential Create------ --------Random
Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read---
-Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec
%CP
                 16    74  97  7394  62  5653  97    78  99   311  90   450
86
tux1,200M,1040,96,3938,13,3244,7,941,86,7269,11,231.1,8,16,74,97,7394,62,5653,97,78,99,311,90,450,86
[zphelj@tux1 bontest]



Jay Phelps
Computer Scientist
CSC Financial Services Group
[EMAIL PROTECTED]
512.275.2150



                    Donna Awald
                    <dawald              To:     [EMAIL PROTECTED]
                    @empagio.com>        cc:
                    Sent by:             Subject:     Re: Problem speed of dasd drives 
on MP 3000
                    Linux on 390
                    Port
                    <LINUX-390


                    04/18/2002
                    08:02 AM
                    Please
                    respond to
                    Linux on 390
                    Port






Tim,
Are you using the emulated IO drives?  or internal disk storage drive.  The
emulated IO drives are really just an OS/2 SCSI PC type drive going thru
alot of software/drivers to emulate a 3380/3390.  We don't use them much
because they have poor performance and high usage of them cause our OS/2
side to lockup, thus locking up the HMC.  We just put developers code on
them.

-----Original Message-----
From: Tim Verhoeven [mailto:[EMAIL PROTECTED]]
Sent: Thursday, April 18, 2002 8:49 AM
To: [EMAIL PROTECTED]
Subject: Re: Problem speed of dasd drives on MP 3000


On Thu, 18 Apr 2002, David Andrews wrote:

> On Thu, 2002-04-18 at 05:02, Tim Verhoeven wrote:
> >
> > So I tested the speed of the Linux partition with bonnie++. I've got a
> > transfer speed of about 5MB/sec. Thats slower then a 100Mb ethernet
link.
>
> What parameters did you feed bonnie?
>
> Here's a run I did in a 128MB partition, 7060-H50, STK 9500
> (Iceberg-like device).  I guessed a 200MB test was sufficient, you may
> know better.
>
> dba@linux:~ > uname -a
> Linux linux 2.2.16 #1 SMP Wed Nov 8 10:57:03 GMT 2000 s390 unknown
> dba@linux:~ > bonnie -s 200
> File './Bonnie.11674', size: 209715200, volumes: 1
> Writing with putc()...  done:   2188 kB/s  98.4 %CPU
> Rewriting...            done:   3745 kB/s   3.5 %CPU
> Writing intelligently...done:   9063 kB/s   6.5 %CPU
> Reading with getc()...  done:   1818 kB/s  90.3 %CPU
> Reading intelligently...done:   9296 kB/s   4.6 %CPU
> Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done...
>               ---Sequential Output (nosync)--- ---Sequential Input--
--Rnd
Seek-
>               -Per Char- --Block--- -Rewrite-- -Per Char- --Block---
--04k
(03)-
> Machine    MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU
/sec %CPU
>        1* 200  2188 98.4  9063  6.5  3745  3.5  1818 90.3  9296  4.6
466.2  3.8

I've run it twice, with a size of 400 and 256. And my figures (input and
output) are about halve of yours.

So there seems to be a problem on my setup, anyone hints how to check, how
to improve ?

Thanks,
Tim

--
============================================================================

==
Tim Verhoeven
                               Music Services - Michel Stoffels
GSM : 0496 / 693 453                          + Deejayteam
Email : [EMAIL PROTECTED]                         + Sound & Light rentals
URL : www.sin.khk.be/~dj/                     + P.A. services
=========Public PGP-Key at :
http://www.sin.khk.be/~dj/publickey.txt==========
          Member of Student Information Networking (www.sin.khk.be)

Reply via email to