onstat -g glo: cpu efficiency
A user worried that onstat -g glo showed only ~65% efficiency on his three CPU VPs. John Miller explained the metric is simply OS process time divided by total thread time on the VP, i.e. an indicator of how well the oninit processes get scheduled by the OS — low values mean threads waited for a hardware CPU. Others noted causes: processor affinity, client apps competing on the same box, over-configuring more CPU VPs than cores, and a known 11.50 bug giving bogus values. The poster found concurrent load/update-statistics jobs on other instances on the same server were colliding, and planned to stagger them; no final outcome is recorded.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
I am a bit concerned over what onstat -g glo is reporting regarding eff for
the three cpuvp's , they are avg in the mid 60's. Other instances in our
client's landscape have all three much closer to 100% (however those instances
are no where near as heavily loaded as this one).
Comments anyone ?
Individual virtual processors:
vp pid class usercpu syscpu total Thread Eff
1 4211134 cpu 21241.47 2534.10 23775.57 36411.97 65%
2 2322448 adm 21.71 20.49 42.20 0.00 0%
3 3444892 cpu 17728.03 2020.35 19748.38 30409.40 64%
4 2908326 cpu 10974.72 1206.09 12180.81 18296.49 66%
5 1396890 lio 0.77 2.78 3.55 3.55 100%
6 4571320 pio 0.79 2.89 3.68 3.68 100%
7 573718 aio 0.98 4.33 5.31 5.31 100%
8 3600816 msc 109.03 84.35 193.38 198.84 97%
9 3272944 fifo 0.77 2.84 3.61 3.61 100%
10 4161568 aio 0.79 2.92 3.71 3.71 100%
11 3764236 soc 423.80 640.31 1064.11 NA NA
tot 50502.86 6521.45 57024.31
Version?...
Overall CPU usage in the machine?
On Tue, Jul 26, 2011 at 10:43 PM, MARK JALKIEWICZ <
mark.jalkiewicz@verizon.net> wrote:
> I am a bit concerned over what onstat -g glo is reporting regarding eff for
> the three cpuvp's , they are avg in the mid 60's. Other instances in our
> client's landscape have all three much closer to 100% (however those
> instances
> are no where near as heavily loaded as this one).
>
> Comments anyone ?
>
> Individual virtual processors:
> vp pid class usercpu syscpu total Thread Eff
> 1 4211134 cpu 21241.47 2534.10 23775.57 36411.97 65%
> 2 2322448 adm 21.71 20.49 42.20 0.00 0%
> 3 3444892 cpu 17728.03 2020.35 19748.38 30409.40 64%
> 4 2908326 cpu 10974.72 1206.09 12180.81 18296.49 66%
> 5 1396890 lio 0.77 2.78 3.55 3.55 100%
> 6 4571320 pio 0.79 2.89 3.68 3.68 100%
> 7 573718 aio 0.98 4.33 5.31 5.31 100%
> 8 3600816 msc 109.03 84.35 193.38 198.84 97%
> 9 3272944 fifo 0.77 2.84 3.61 3.61 100%
> 10 4161568 aio 0.79 2.92 3.71 3.71 100%
> 11 3764236 soc 423.80 640.31 1064.11 NA NA
>
> tot 50502.86 6521.45 57024.31
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...
--0016368e23a5ec72dd04a9002dbe
The efficiency is an indicator of how well the oninits are being scheduled
by the operating system.
In short the formula behind the seens is very simple ratio of
( Total OS process time ) / ( total time of the informix threads on
the vps)
In the stats you provided below:
VP #1
=====
Total time given to the processor
23775.57
Total time of all threads running on the VP 36411.97
This means that threads where running on the informix oninit process
for (36,411 - 23775 = ) 12,636 seconds without finding a hardware processor
to run on.
John F. Miller III
STSM, Embedability Architect
miller3@us.ibm.com
503-578-5645
IBM Informix Dynamic Server (IDS)
ids-bounces@iiug.org wrote on 07/26/2011 02:43:03 PM:
> From: "MARK JALKIEWICZ" <mark.jalkiewicz@verizon.net>
> To: ids@iiug.org
> Date: 07/26/2011 02:43 PM
> Subject: onstat -g glo: cpu efficiency [24459]
> Sent by: ids-bounces@iiug.org
>
> I am a bit concerned over what onstat -g glo is reporting regarding eff
for
> the three cpuvp's , they are avg in the mid 60's. Other instances in our
> client's landscape have all three much closer to 100% (however
thoseinstances
> are no where near as heavily loaded as this one).
>
> Comments anyone ?
>
> Individual virtual processors:
> vp pid class usercpu syscpu total Thread Eff
> 1 4211134 cpu 21241.47 2534.10 23775.57 36411.97 65%
> 2 2322448 adm 21.71 20.49 42.20 0.00 0%
> 3 3444892 cpu 17728.03 2020.35 19748.38 30409.40 64%
> 4 2908326 cpu 10974.72 1206.09 12180.81 18296.49 66%
> 5 1396890 lio 0.77 2.78 3.55 3.55 100%
> 6 4571320 pio 0.79 2.89 3.68 3.68 100%
> 7 573718 aio 0.98 4.33 5.31 5.31 100%
> 8 3600816 msc 109.03 84.35 193.38 198.84 97%
> 9 3272944 fifo 0.77 2.84 3.61 3.61 100%
> 10 4161568 aio 0.79 2.92 3.71 3.71 100%
> 11 3764236 soc 423.80 640.31 1064.11 NA NA
>
> tot 50502.86 6521.45 57024.31
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
I've seen low efficiency numbers mostly when affinity is used.
Art
Art S. Kagel
Advanced DataTools (www.advancedatatools.com)
Blog: http://informix-myview.blogspot.com/
Disclaimer: Please keep in mind that my own opinions are my own opinions and
do not reflect on my employer, Advanced DataTools, the IIUG, nor any other
organization with which I am associated either explicitly, implicitly, or by
inference. Neither do those opinions reflect those of other individuals
affiliated with any entity with which I am affiliated nor those of the
entities themselves.
On Tue, Jul 26, 2011 at 8:03 PM, John Miller iii <miller3@us.ibm.com> wrote:
> The efficiency is an indicator of how well the oninits are being scheduled
> by the operating system.
>
> In short the formula behind the seens is very simple ratio of
>
> ( Total OS process time ) / ( total time of the informix threads on
> the vps)
>
> In the stats you provided below:
>
> VP #1
> =====
> Total time given to the processor
> 23775.57
> Total time of all threads running on the VP 36411.97
>
> This means that threads where running on the informix oninit process
> for (36,411 - 23775 = ) 12,636 seconds without finding a hardware processor
> to run on.
>
> John F. Miller III
> STSM, Embedability Architect
> miller3@us.ibm.com
> 503-578-5645
> IBM Informix Dynamic Server (IDS)
>
> ids-bounces@iiug.org wrote on 07/26/2011 02:43:03 PM:
>
> > From: "MARK JALKIEWICZ" <mark.jalkiewicz@verizon.net>
> > To: ids@iiug.org
> > Date: 07/26/2011 02:43 PM
> > Subject: onstat -g glo: cpu efficiency [24459]
> > Sent by: ids-bounces@iiug.org
> >
> > I am a bit concerned over what onstat -g glo is reporting regarding eff
> for
> > the three cpuvp's , they are avg in the mid 60's. Other instances in our
> > client's landscape have all three much closer to 100% (however
> thoseinstances
> > are no where near as heavily loaded as this one).
> >
> > Comments anyone ?
> >
> > Individual virtual processors:
> > vp pid class usercpu syscpu total Thread Eff
> > 1 4211134 cpu 21241.47 2534.10 23775.57 36411.97 65%
> > 2 2322448 adm 21.71 20.49 42.20 0.00 0%
> > 3 3444892 cpu 17728.03 2020.35 19748.38 30409.40 64%
> > 4 2908326 cpu 10974.72 1206.09 12180.81 18296.49 66%
> > 5 1396890 lio 0.77 2.78 3.55 3.55 100%
> > 6 4571320 pio 0.79 2.89 3.68 3.68 100%
> > 7 573718 aio 0.98 4.33 5.31 5.31 100%
> > 8 3600816 msc 109.03 84.35 193.38 198.84 97%
> > 9 3272944 fifo 0.77 2.84 3.61 3.61 100%
> > 10 4161568 aio 0.79 2.92 3.71 3.71 100%
> > 11 3764236 soc 423.80 640.31 1064.11 NA NA
> >
> > tot 50502.86 6521.45 57024.31
> >
> >
> >
>
>
>
*******************************************************************************
>
> > Forum Note: Use "Reply" to post a response in the discussion forum.
> >
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--20cf307d00f06a453904a901df34
This is interesting, just looking at a couple of our systems...
Efficiency on CPU threads sits between 60-80%. I suspect that's probably about
right, because in our environment the client application is on the same box
competing for resources...
The efficiency on all the io classes, PIO/LIO/AIO is very low, between 0 and
3%. Is this because the kernel de-schedules these threads while waiting on the
actual io to complete?
Thanks,
Tristan
Tristan Ball - Hosted Services Manager VIC
Pronto Hosted Services
20 Lakeside Drive, Burwood East, VIC 3151
Phone: +61 3 9887 7770 | Email: tristanb@pronto.com.au
Mobile: +61 408 397 473
For PHS helpdesk support, please email phs@pronto.com.au
For urgent after hours support phone: 1800 622 556
---Legal Notice---
The email message and any attachments are confidential and subject to
copyright. If you are not the intended recipient, any use, interference with,
disclosure or copying of this material is unauthorised and prohibited. No part
may be reproduced, adapted or transmitted without the written permission of
the copyright owner. If you have received this email in error, please
immediately advise the sender by return email and delete the message from your
system. Before opening or using attachments, check for viruses and defects.
Our liability is limited to re-supplying any affected attachments.
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Art Kagel
Sent: Wednesday, 27 July 2011 10:10 AM
To: ids@iiug.org
Subject: Re: onstat -g glo: cpu efficiency [24462]
I've seen low efficiency numbers mostly when affinity is used.
Art
Art S. Kagel
Advanced DataTools (www.advancedatatools.com)
Blog: http://informix-myview.blogspot.com/
Disclaimer: Please keep in mind that my own opinions are my own opinions and
do not reflect on my employer, Advanced DataTools, the IIUG, nor any other
organization with which I am associated either explicitly, implicitly, or by
inference. Neither do those opinions reflect those of other individuals
affiliated with any entity with which I am affiliated nor those of the
entities themselves.
On Tue, Jul 26, 2011 at 8:03 PM, John Miller iii <miller3@us.ibm.com> wrote:
> The efficiency is an indicator of how well the oninits are being
> scheduled by the operating system.
>
> In short the formula behind the seens is very simple ratio of
>
> ( Total OS process time ) / ( total time of the informix threads on
> the vps)
>
> In the stats you provided below:
>
> VP #1
> =====
> Total time given to the processor
> 23775.57
> Total time of all threads running on the VP 36411.97
>
> This means that threads where running on the informix oninit process
> for (36,411 - 23775 = ) 12,636 seconds without finding a hardware
> processor to run on.
>
> John F. Miller III
> STSM, Embedability Architect
> miller3@us.ibm.com
> 503-578-5645
> IBM Informix Dynamic Server (IDS)
>
> ids-bounces@iiug.org wrote on 07/26/2011 02:43:03 PM:
>
> > From: "MARK JALKIEWICZ" <mark.jalkiewicz@verizon.net>
> > To: ids@iiug.org
> > Date: 07/26/2011 02:43 PM
> > Subject: onstat -g glo: cpu efficiency [24459] Sent by:
> > ids-bounces@iiug.org
> >
> > I am a bit concerned over what onstat -g glo is reporting regarding
> > eff
> for
> > the three cpuvp's , they are avg in the mid 60's. Other instances in
> > our client's landscape have all three much closer to 100% (however
> thoseinstances
> > are no where near as heavily loaded as this one).
> >
> > Comments anyone ?
> >
> > Individual virtual processors:
> > vp pid class usercpu syscpu total Thread Eff
> > 1 4211134 cpu 21241.47 2534.10 23775.57 36411.97 65%
> > 2 2322448 adm 21.71 20.49 42.20 0.00 0%
> > 3 3444892 cpu 17728.03 2020.35 19748.38 30409.40 64%
> > 4 2908326 cpu 10974.72 1206.09 12180.81 18296.49 66%
> > 5 1396890 lio 0.77 2.78 3.55 3.55 100%
> > 6 4571320 pio 0.79 2.89 3.68 3.68 100%
> > 7 573718 aio 0.98 4.33 5.31 5.31 100%
> > 8 3600816 msc 109.03 84.35 193.38 198.84 97%
> > 9 3272944 fifo 0.77 2.84 3.61 3.61 100%
> > 10 4161568 aio 0.79 2.92 3.71 3.71 100%
> > 11 3764236 soc 423.80 640.31 1064.11 NA NA
> >
> > tot 50502.86 6521.45 57024.31
> >
> >
> >
>
>
>
*******************************************************************************
>
> > Forum Note: Use "Reply" to post a response in the discussion forum.
> >
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--20cf307d00f06a453904a901df34
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
We have two rather similar systems. They both have 16 cores. One we have
configured with 15 CPU VP's. The other is configured with 25 CPU VP's.
The first system shows an efficiency between 68 and 80, the second not
more than 55. Seems to me that "overbooking" of the real CPU's decreases
the efficiency of a single CPU VP.
Not really astonishing, though ;-)
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
Tristan
> Ball
> Sent: Wednesday, July 27, 2011 6:45 AM
> To: ids@iiug.org
> Subject: RE: onstat -g glo: cpu efficiency [24463]
>
> This is interesting, just looking at a couple of our systems...
>
> Efficiency on CPU threads sits between 60-80%. I suspect that's
probably about
> right, because in our environment the client application is on the
same box
> competing for resources...
>
> The efficiency on all the io classes, PIO/LIO/AIO is very low, between
0 and
> 3%. Is this because the kernel de-schedules these threads while
waiting on the
> actual io to complete?
>
> Thanks,
>
> Tristan
>
> Tristan Ball - Hosted Services Manager VIC
> Pronto Hosted Services
> 20 Lakeside Drive, Burwood East, VIC 3151
> Phone: +61 3 9887 7770 | Email: tristanb@pronto.com.au
> Mobile: +61 408 397 473
>
> For PHS helpdesk support, please email phs@pronto.com.au
> For urgent after hours support phone: 1800 622 556
>
> ---Legal Notice---
> The email message and any attachments are confidential and subject to
> copyright. If you are not the intended recipient, any use,
interference with,
> disclosure or copying of this material is unauthorised and prohibited.
No part
> may be reproduced, adapted or transmitted without the written
permission of
> the copyright owner. If you have received this email in error, please
> immediately advise the sender by return email and delete the message
from your
> system. Before opening or using attachments, check for viruses and
defects.
> Our liability is limited to re-supplying any affected attachments.
>
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
Art
> Kagel
> Sent: Wednesday, 27 July 2011 10:10 AM
> To: ids@iiug.org
> Subject: Re: onstat -g glo: cpu efficiency [24462]
>
> I've seen low efficiency numbers mostly when affinity is used.
>
> Art
>
> Art S. Kagel
> Advanced DataTools (www.advancedatatools.com)
> Blog: http://informix-myview.blogspot.com/
>
> Disclaimer: Please keep in mind that my own opinions are my own
opinions and
> do not reflect on my employer, Advanced DataTools, the IIUG, nor any
other
> organization with which I am associated either explicitly, implicitly,
or by
> inference. Neither do those opinions reflect those of other
individuals
> affiliated with any entity with which I am affiliated nor those of the
> entities themselves.
>
> On Tue, Jul 26, 2011 at 8:03 PM, John Miller iii <miller3@us.ibm.com>
wrote:
>
> > The efficiency is an indicator of how well the oninits are being
> > scheduled by the operating system.
> >
> > In short the formula behind the seens is very simple ratio of
> >
> > ( Total OS process time ) / ( total time of the informix threads on
> > the vps)
> >
> > In the stats you provided below:
> >
> > VP #1
> > =====
> > Total time given to the processor
> > 23775.57
> > Total time of all threads running on the VP 36411.97
> >
> > This means that threads where running on the informix oninit process
> > for (36,411 - 23775 = ) 12,636 seconds without finding a hardware
> > processor to run on.
> >
> > John F. Miller III
> > STSM, Embedability Architect
> > miller3@us.ibm.com
> > 503-578-5645
> > IBM Informix Dynamic Server (IDS)
> >
> > ids-bounces@iiug.org wrote on 07/26/2011 02:43:03 PM:
> >
> > > From: "MARK JALKIEWICZ" <mark.jalkiewicz@verizon.net>
> > > To: ids@iiug.org
> > > Date: 07/26/2011 02:43 PM
> > > Subject: onstat -g glo: cpu efficiency [24459] Sent by:
> > > ids-bounces@iiug.org
> > >
> > > I am a bit concerned over what onstat -g glo is reporting
regarding
> > > eff
> > for
> > > the three cpuvp's , they are avg in the mid 60's. Other instances
in
> > > our client's landscape have all three much closer to 100% (however
> > thoseinstances
> > > are no where near as heavily loaded as this one).
> > >
> > > Comments anyone ?
> > >
> > > Individual virtual processors:
> > > vp pid class usercpu syscpu total Thread Eff
> > > 1 4211134 cpu 21241.47 2534.10 23775.57 36411.97 65%
> > > 2 2322448 adm 21.71 20.49 42.20 0.00 0%
> > > 3 3444892 cpu 17728.03 2020.35 19748.38 30409.40 64%
> > > 4 2908326 cpu 10974.72 1206.09 12180.81 18296.49 66%
> > > 5 1396890 lio 0.77 2.78 3.55 3.55 100%
> > > 6 4571320 pio 0.79 2.89 3.68 3.68 100%
> > > 7 573718 aio 0.98 4.33 5.31 5.31 100%
> > > 8 3600816 msc 109.03 84.35 193.38 198.84 97%
> > > 9 3272944 fifo 0.77 2.84 3.61 3.61 100%
> > > 10 4161568 aio 0.79 2.92 3.71 3.71 100%
> > > 11 3764236 soc 423.80 640.31 1064.11 NA NA
> > >
> > > tot 50502.86 6521.45 57024.31
> > >
> > >
> > >
> >
> >
> >
>
>
************************************************************************
*******
> >
> > > Forum Note: Use "Reply" to post a response in the discussion
forum.
> > >
> >
> >
> >
> >
>
>
************************************************************************
*******
> > Forum Note: Use "Reply" to post a response in the discussion forum.
> >
> >
>
> --20cf307d00f06a453904a901df34
>
>
>
************************************************************************
*******
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
>
************************************************************************
*******
> Forum Note: Use "Reply" to post a response in the discussion forum.
Also note that there was a bug in 11.50 that lead to wrong values.
I can't recall the details but I think the shown values were extremely low
(not like the values mentioned here). I also can't remember the version, but
I think it should be around FC3-FC5.
If needed I can search the details.
Regards.
On Wed, Jul 27, 2011 at 7:57 AM, Habichtsberg, Reinhard <
RHabichtsberg@arz-emmendingen.de> wrote:
> We have two rather similar systems. They both have 16 cores. One we have
> configured with 15 CPU VP's. The other is configured with 25 CPU VP's.
>
> The first system shows an efficiency between 68 and 80, the second not
> more than 55. Seems to me that "overbooking" of the real CPU's decreases
> the efficiency of a single CPU VP.
>
> Not really astonishing, though ;-)
>
> > -----Original Message-----
> > From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
> Tristan
> > Ball
> > Sent: Wednesday, July 27, 2011 6:45 AM
> > To: ids@iiug.org
> > Subject: RE: onstat -g glo: cpu efficiency [24463]
> >
> > This is interesting, just looking at a couple of our systems...
> >
> > Efficiency on CPU threads sits between 60-80%. I suspect that's
> probably about
> > right, because in our environment the client application is on the
> same box
> > competing for resources...
> >
> > The efficiency on all the io classes, PIO/LIO/AIO is very low, between
> 0 and
> > 3%. Is this because the kernel de-schedules these threads while
> waiting on the
> > actual io to complete?
> >
> > Thanks,
> >
> > Tristan
> >
> > Tristan Ball - Hosted Services Manager VIC
> > Pronto Hosted Services
> > 20 Lakeside Drive, Burwood East, VIC 3151
> > Phone: +61 3 9887 7770 | Email: tristanb@pronto.com.au
> > Mobile: +61 408 397 473
> >
> > For PHS helpdesk support, please email phs@pronto.com.au
> > For urgent after hours support phone: 1800 622 556
> >
> > ---Legal Notice---
> > The email message and any attachments are confidential and subject to
> > copyright. If you are not the intended recipient, any use,
> interference with,
> > disclosure or copying of this material is unauthorised and prohibited.
> No part
> > may be reproduced, adapted or transmitted without the written
> permission of
> > the copyright owner. If you have received this email in error, please
> > immediately advise the sender by return email and delete the message
> from your
> > system. Before opening or using attachments, check for viruses and
> defects.
> > Our liability is limited to re-supplying any affected attachments.
> >
> > -----Original Message-----
> > From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
> Art
> > Kagel
> > Sent: Wednesday, 27 July 2011 10:10 AM
> > To: ids@iiug.org
> > Subject: Re: onstat -g glo: cpu efficiency [24462]
> >
> > I've seen low efficiency numbers mostly when affinity is used.
> >
> > Art
> >
> > Art S. Kagel
> > Advanced DataTools (www.advancedatatools.com)
> > Blog: http://informix-myview.blogspot.com/
> >
> > Disclaimer: Please keep in mind that my own opinions are my own
> opinions and
> > do not reflect on my employer, Advanced DataTools, the IIUG, nor any
> other
> > organization with which I am associated either explicitly, implicitly,
> or by
> > inference. Neither do those opinions reflect those of other
> individuals
> > affiliated with any entity with which I am affiliated nor those of the
> > entities themselves.
> >
> > On Tue, Jul 26, 2011 at 8:03 PM, John Miller iii <miller3@us.ibm.com>
> wrote:
> >
> > > The efficiency is an indicator of how well the oninits are being
> > > scheduled by the operating system.
> > >
> > > In short the formula behind the seens is very simple ratio of
> > >
> > > ( Total OS process time ) / ( total time of the informix threads on
> > > the vps)
> > >
> > > In the stats you provided below:
> > >
> > > VP #1
> > > =====
> > > Total time given to the processor
> > > 23775.57
> > > Total time of all threads running on the VP 36411.97
> > >
> > > This means that threads where running on the informix oninit process
> > > for (36,411 - 23775 = ) 12,636 seconds without finding a hardware
> > > processor to run on.
> > >
> > > John F. Miller III
> > > STSM, Embedability Architect
> > > miller3@us.ibm.com
> > > 503-578-5645
> > > IBM Informix Dynamic Server (IDS)
> > >
> > > ids-bounces@iiug.org wrote on 07/26/2011 02:43:03 PM:
> > >
> > > > From: "MARK JALKIEWICZ" <mark.jalkiewicz@verizon.net>
> > > > To: ids@iiug.org
> > > > Date: 07/26/2011 02:43 PM
> > > > Subject: onstat -g glo: cpu efficiency [24459] Sent by:
> > > > ids-bounces@iiug.org
> > > >
> > > > I am a bit concerned over what onstat -g glo is reporting
> regarding
> > > > eff
> > > for
> > > > the three cpuvp's , they are avg in the mid 60's. Other instances
> in
> > > > our client's landscape have all three much closer to 100% (however
> > > thoseinstances
> > > > are no where near as heavily loaded as this one).
> > > >
> > > > Comments anyone ?
> > > >
> > > > Individual virtual processors:
> > > > vp pid class usercpu syscpu total Thread Eff
> > > > 1 4211134 cpu 21241.47 2534.10 23775.57 36411.97 65%
> > > > 2 2322448 adm 21.71 20.49 42.20 0.00 0%
> > > > 3 3444892 cpu 17728.03 2020.35 19748.38 30409.40 64%
> > > > 4 2908326 cpu 10974.72 1206.09 12180.81 18296.49 66%
> > > > 5 1396890 lio 0.77 2.78 3.55 3.55 100%
> > > > 6 4571320 pio 0.79 2.89 3.68 3.68 100%
> > > > 7 573718 aio 0.98 4.33 5.31 5.31 100%
> > > > 8 3600816 msc 109.03 84.35 193.38 198.84 97%
> > > > 9 3272944 fifo 0.77 2.84 3.61 3.61 100%
> > > > 10 4161568 aio 0.79 2.92 3.71 3.71 100%
> > > > 11 3764236 soc 423.80 640.31 1064.11 NA NA
> > > >
> > > > tot 50502.86 6521.45 57024.31
> > > >
> > > >
> > > >
> > >
> > >
> > >
> >
> >
> ************************************************************************
> *******
> > >
> > > > Forum Note: Use "Reply" to post a response in the discussion
> forum.
> > > >
> > >
> > >
> > >
> > >
> >
> >
> ************************************************************************
> *******
> > > Forum Note: Use "Reply" to post a response in the discussion forum.
> > >
> > >
> >
> > --20cf307d00f06a453904a901df34
> >
> >
> >
> ************************************************************************
> *******
> > Forum Note: Use "Reply" to post a response in the discussion forum.
> >
> >
> >
> ************************************************************************
> *******
> > Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...@@NL@
I did some more digging and found that every n minutes load jobs on competing instances (same server) wake up at the same time and load data into tables and then run update statistics. The data loads are not what you can consider large , but just enough to cause some contention I presume. Next action is to try and seperate those wakup calls.
Related threads
- Posting from the Informix-list
- Migrating from IDS 9.40.UC6 to 11.50.UC3
- Ip for a network session
- questions onstat -g