This is my goals:
/**********************************************************************/
/* ADMIN STATEMENT */
/* This will cause messages to be sent to VMRMADMN's console */
ADMIN MSGUSER OPERATOR
/* GOAL STATEMENTS */
GOAL MAX VELOCITY CPU 90 DASD 90
GOAL MIDCPU VELOCITY CPU 60
GOAL MINCPU VELOCITY CPU 10
/* WORKLOAD statements followed by corresponding MANAGE statement */
* workload 1
WORKLOAD WORK1 USER SESAME
MANAGE WORK1 GOAL MAX IMPORTANCE 10
* workload 2
WORKLOAD WORK2 USER SCORTPRO
MANAGE WORK2 GOAL MAX IMPORTANCE 8
* workload 3
WORKLOAD WORK3 USER FRPRO LINUXP1
MANAGE WORK3 GOAL MIDCPU IMPORTANCE 6
* WORKLOAD 4
WORKLOAD WORK4 USER SESPRE SESINT SESINT8 FRPRE WASADM SCORTPRE
MANAGE WORK4 GOAL MIDCPU IMPORTANCE 2
* WORKLOAD 5
WORKLOAD WORK5 USER LNXTRS7 XAVIER LINSVG LNXTRS8
MANAGE WORK5 GOAL MINCPU IMPORTANCE 1
and this is what I get:
<-Seconds-> T/V
Userid %CPU TCPU VCPU Ratio Total DASD Avoid Diag98 U
>System< 3.09 1.852 1.814 1.0 1.0 1.0 .2 .0 .
FRPRO 31.0 18.63 18.36 1.0 2.7 2.7 .0 .0 .
SCORTPRE 28.2 16.92 16.71 1.0 4.9 4.9 2.9 .0 .
LINUXP1 5.36 3.217 3.129 1.0 2.1 2.1 .0 .0 .
FRPRE 2.11 1.268 1.232 1.0 .3 .3 .0 .0 .
SCORTPRO .99 .592 .578 1.0 1.7 1.7 .0 .0 .
SESPRE .92 .550 .531 1.0 2.6 2.6 .0 .0 .
SESINT .69 .413 .394 1.0 2.7 2.7 .0 .0 .
SESINT8 .66 .395 .381 1.0 2.6 2.6 .0 .0 .
SESAME .41 .248 .237 1.0 1.6 1.6 .0 .0 .
TCPIP .29 .176 .002 88.0 .0 .0 .0 .0 .
FIREW .23 .138 .099 1.4 .0 .0 .0 .0 .
VMRMSVM .07 .041 .039 1.1 1.0 1.0 .7 .0 .
PERFSVM .02 .014 .012 1.2 .1 .0 .0 .0 .
LINSVG .01 .008 .006 1.3 .0 .0 .0 .0 .
EREP 0 0 0 ... 0 0 0 0
> <--Minutes--> Total Rate Insert
Userid > UR Pg/s User Status Logged Active Pages SPg/s MDC/s Share
>System< > .0 .0 ---,---,---- 1.0 .7 .0 .0 .0 ---
FRPRO .0 .0 ESA,CL0,DISP 1 1 0 .00 .0 2
LINUXP1 .0 .0 ESA,CL1,DISP 1 1 0 .00 .0 204
FRPRE .0 .0 ESA,CL3,DISP 1 1 0 .00 .0 407
SCORTPRE .0 .0 ESA,CL2,DISP 1 1 0 .00 .0 848
SCORTPRO .0 .4 ESA,CL0,DISP 1 1 0 .00 .0 4638
SESPRE .0 .1 ESA,CL3,DISP 1 1 0 .00 .0 2880
SESINT8 .0 .0 ESA,CL2,DISP 1 1 0 .00 .0 180
SESINT .0 .0 ESA,CL2,DISP 1 1 0 .00 .0 8372
SESAME .0 .0 ESA,CL0,DISP 1 1 0 .00 .0 10000
TCPIP .0 .0 ESA,CL0,DISP 1 1 0 .00 .0 3000
VMRMSVM .0 .0 ESA,---,DORM 1 1 0 .00 .0 3.0%A
FIREW .0 .0 ESA,CL3,DISP 1 1 0 .00 .0 100
PERFSVM .0 .0 ESA,---,DORM 1 1 0 .00 .0 3.0%A
LINSVG .0 .0 ESA,CL3,DISP 1 1 0 .00 .0 18
VMSERVS .0 .0 XC, ---,DORM 1 1 0 .00 .0 1500
You can see that "share values" are not set correctly.
If I do a loop on SCORTPRE for example, consummations are near 150% and
FRPRO don't respond any more. SCORTPRE is a development guest, FRPRO a
production guest, I would like to give CPU resource to SCORTPRE only if
FRPRO don't need CPU resource. The priority should be FRPRO. If FRPRO
don't need CPU resource, SCORTPRE could keep all resource available.
Thanks for your help.
Gerard MONTELEONE
-----Message d'origine-----
De : Linux on 390 Port [mailto:[EMAIL PROTECTED] De la part de
David Boyes
Envoyé : mercredi 15 septembre 2004 02:18
À : [EMAIL PROTECTED]
Objet : Re: VMRMSVM Z/VM 1.4
On Wed, Sep 14, 2005 at 05:36:38PM +0200, Monteleone wrote:
> Has somebody already get success with VMRMSVM ?
> I'm not able to obtain the performance goals I defined.
> Is there another way to dynamically adjust CPU share and I.O ?
You could use the VM Perfkit and define thresholds for specific users,
but I don't know if it's going to help you or not, since I don't know
what you're trying to do. You may simply not have enough resources to
meet your goals.
What kind of goals are you trying to accomplish?
-- db
----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or
visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
