CPU utilization on Solaris

When running a series of performance tests on Solaris machine running Weblogic
8.1sp2 and comparing the results to a Windows200 machine running Weblogic 8.1sp2
I noticed the Solaris box was only using 10% of the each of its 4 CPUs. The Window2k
machine was however using 85% of it's single CPU and was performing 3 time better
than the Solaris machine.
I was hoping someone might have some suggestions for where to look for causes
of the cpus being under utilized.
Thanks
-Fazle

Similar Messages

Syslogd Consuming more CPU utilization in Solaris 10

Hi All,
The Syslogd process consuming more CPU utilization in Solaris 10. Kindly help how to reduce this cpu utilization.
Regards
Siva

Hi Robert,
Both are same architecture. x86
The following is the prstat o/p from the affected server. Pls note that one of the mount point in this server is in ZFS.
prstat -l
PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/LWPID
26092 root 3933M 3930M cpu1 60 0 29:00:43 22% syslogd/56
26092 root 3933M 3930M run 30 0 289:47:33 12% syslogd/18
26092 root 3933M 3930M run 40 0 272:31:05 11% syslogd/22
26092 root 3933M 3930M run 22 0 14:47:16 9.7% syslogd/65
26092 root 3933M 3930M run 42 0 14:43:46 9.7% syslogd/63
26092 root 3933M 3930M run 31 0 14:40:42 9.6% syslogd/66
26092 root 3933M 3930M sleep 40 0 152:45:42 5.9% syslogd/21
26092 root 3933M 3930M cpu0 53 0 6:41:58 4.1% syslogd/58
26092 root 3933M 3930M run 52 0 6:23:13 4.0% syslogd/57
26092 root 3933M 3930M sleep 43 0 6:29:21 3.9% syslogd/59
26092 root 3933M 3930M sleep 52 0 5:55:14 3.6% syslogd/71
More over we are continuously receiving the below error message in the /var/adm/messages, we don't know, from where the error arises from the syslog server.
syslogd: malloc failed: dropping message from remote: Not enough space
PRIVILEGE :[4] 'NONE'
Edited by: Siva_Systems on Mar 29, 2010 5:06 AM
Edited by: Siva_Systems on Mar 29, 2010 8:18 AM

Xsun (high CPU utilization) on Solaris 10 Sparc

hi
i have sun blade 1500 and am running solaris 10 on it. the machine is a 2 CPU (750Mhz) 4GB Sparc with the lastest cluster patch.
The Xsun process is alway at 50% util and the windowmaker (wmaker) is at 27%.
The Xsun is alway using all available CPU and the machine is really slow.. any help on what patch will fix the Xsun process ? any operation on the machine increases the Xsun's CPU util.
thx
Sriram

Hi sridhar,
can i know which platform you are using....
if it is solaris,
can you paste the details of the prstat -L -p wlpid 1 1 ---> which gives the light wieght pid threads
and also (pstact wlspid) for lwpid to process id mapping
or you can follow these steps for finding which thread is causing the hight cpu utilization
1.find the highest usage lwpid in prstat output
2.find the lwpid in pstack output and get the matching thread number
3.convert the thread number to hexadecimal
4.find the hexadecimal thread number in the server thread dump (nid= xxx)
5.determine what thread was doing to cause the high cpu usage
you can find similar way if it is linux..
ps u -Lp wlspid and thread dump
Thank you,
Bob
Edited by: Bob on Sep 21, 2010 10:18 AM
Edited by: Bob on Sep 21, 2010 10:24 AM

Low CPU utilization on Solaris

Hi all.
We've recently been performance tuning our java application running
inside of an Application Server with Java 1.3.1 Hotspot -server. We've
begun to notice some odd trends and were curious if anyone else out
there has seen similiar things.
Performance numbers show that our server runs twice
as fast on Intel with Win2K than on an Ultra60 with Solaris 2.8.
Here's the hardware information:
Intel -> 2 processors (32bit) at 867 MHz and 2 Gig RAM
Solaris -> 2 processors (64bit) at 450 MHz and 2 Gig RAM.
Throughput for most use cases in a low number of threads is twice as
fast on Intel. The only exception is some of our use-cases that are
heavily dependent on a stored procedure which runs twice as fast on
Solaris. The database (oracle 8i) and the app server run on the same
machine in these tests.
There should minor (or no) network traffic. GC does not seem to be an
issue. We set the max heap at 1024 MG. We tried the various solaris
threading models as recommended, but they have accomplished little.
It is possible our Solaris machine is not configured properly in some
way.
My question (after all that ...) is whether this seems normal to
anyone? Should throughput be higher since the processors are faster on
the wIntel box? Does the fact that the solaris processors are 64bit
have any benefit?
We have also run the HeapTest recommended on this site on both
machines. We found that the memory test performs twice as fast on
solaris, but the CPU test performs 4 times as slow on solaris. The
"joint" test performs twice as slow on solaris. Does this imply bad
things about our solaris configuration? Or is this a normal result?
Another big difference is between Solaris and Win2K in these runs is
that CPU Utilization is low on solaris (20-30%) while its much higher
on Win2K (60-70%)
[both machines are 2 processor and the tests are "primarily" single
threaded at
this stage]. I would except the solaris CPU utilization to be around
50% as well. Any ideas why it isn't?

Hi,
I recently went down this path and wound up coming to the realization that the
cpu's are almost neck and neck per cycle when running my Java app. Let me qualify
this a little more (400mhz Sparc II cpu vs 500mhz Intel cpu) under similar load
running the same test gave me similar results. It wasn't as huge difference in
performance as I was expecting.
My theory is given the scalability of the SPARC architecture, more chips==more
performance with less hardware, whereas the Wintel boxes are cheaper, but in order
to get scaling, the underlying hardware comes into question. (how many wintel
boxes to cluster, co-locate, manage, etc…)
From what little I've found out when running tests against our Solaris 8 (E-250's)
400mhz UltraSparc 2's is that it appears that the CPU performance in a lightly
threaded environment is almost 1 cycle / 1 cycle (SPARC to Intel). I don't think
the 64 bit SPARC architecture will buy you anything for java 1.3.1, but if your
application has some huge memory requirements, then using 1.4.0(when BEA supports
it) should be beneficial (check out http://java.sun.com/j2se/1.4/performance.guide.html).
If your application is running only a few threads, tying the threads to the LWP
kernel processes probably won't gain you much. I noticed that it decreased performance
for a test with only a few threads.
I can't give you a good reason as to why your Solaris CPU utilization is so low,
you may want to try getting a copy of Jprobe and profiling Weblogic and your application
to see where your bottlenecks are. I was able to do this with our product, and
found some nasty little performance bugs, but even with that our CPU utilization
was around 98% on a single and 50% on a dual.
Also, take a look at iostat / vmstat and see if your system is bottlenecking doing
io operations. I kept a background process of vmstat to a log and then looked
at it after my test and saw that my cpu was constantly pegged out (doing a lot
of context switching), but that it wasn't doing a whole lot of page faults
(had enough memory).
If you're doing a lot of serialization, that could explain slow performance as
well.
I did follow a suggestion on this board of running my test several times with
the optimizer (-server) and it boosted performance on each iteration until a plateau
on or about the 3rd test.
If you're running Oracle or another RDBMS on your Solaris machine you should see
a pretty decent performance benchmark against NT as these types of applications
are more geared toward the SPARC architecture. From what I've seen running Oracle
on Solaris is pretty darn fast when compared to Intel.
I know that I tried a lot of different tweaks on my Solaris configuration (tcp
buffer size, etc/system parameters for file descriptors, etc.) I even got to the
point where I wanted
to see how WebLogic was handling the Nagle algorithm as far as it's POSIX muxer
was concerned and ran a little test to see how they were setting the sockets (setTcpNoDelay(Boolean)
on java.net.Socket). They're disabling the Nagle algorithm so that wasn't an
issue sigh. My best advice would be to profile your application and see where
the bottlenecks are, you might be able to increase performance, but I'm not too
sure. I also checked out www.spec.org and saw some of their benchmarks that
coincide with our findings.
Best of luck to you and I hope this helps :)
Andy
[email protected] (feanor73) wrote:
Hi all.
We've recently been performance tuning our java application running
inside of an Application Server with Java 1.3.1 Hotspot -server. We've
begun to notice some odd trends and were curious if anyone else out
there has seen similiar things.
Performance numbers show that our server runs twice
as fast on Intel with Win2K than on an Ultra60 with Solaris 2.8.
Here's the hardware information:
Intel -> 2 processors (32bit) at 867 MHz and 2 Gig RAM
Solaris -> 2 processors (64bit) at 450 MHz and 2 Gig RAM.
Throughput for most use cases in a low number of threads is twice as
fast on Intel. The only exception is some of our use-cases that are
heavily dependent on a stored procedure which runs twice as fast on
Solaris. The database (oracle 8i) and the app server run on the same
machine in these tests.
There should minor (or no) network traffic. GC does not seem to be an
issue. We set the max heap at 1024 MG. We tried the various solaris
threading models as recommended, but they have accomplished little.
It is possible our Solaris machine is not configured properly in some
way.
My question (after all that ...) is whether this seems normal to
anyone? Should throughput be higher since the processors are faster on
the wIntel box? Does the fact that the solaris processors are 64bit
have any benefit?
We have also run the HeapTest recommended on this site on both
machines. We found that the memory test performs twice as fast on
solaris, but the CPU test performs 4 times as slow on solaris. The
"joint" test performs twice as slow on solaris. Does this imply bad
things about our solaris configuration? Or is this a normal result?
Another big difference is between Solaris and Win2K in these runs is
that CPU Utilization is low on solaris (20-30%) while its much higher
on Win2K (60-70%)
[both machines are 2 processor and the tests are "primarily" single
threaded at
this stage]. I would except the solaris CPU utilization to be around
50% as well. Any ideas why it isn't?

HELP: WL 8.1 run unexpectedly slow on solaris 9 with low CPU utilization

Hi All
I have setup my app to run on WL8.1 + solaris 9 env + SUN mid-end series server.
JVM was configured to use 3G RAM and there are still abundant RAM on the HW server.
I tried out a use case, it took long time to get response (~ 2 minutes). But the
CPU utilization has been always lower 20%. I have tried out the same test case
on a wintel server with 500 RAM allocated to JVM, the response time is much quicker
(less than 30 sec). I did the same on solaris 8 with 3G RAM and had used alternate
threads library mode (changing LD_LIBRARY_PATH to include /usr/lib/lwp) which
is the default mode adopted by solaris 9. The same use case responded much quicker
and comparable to abovementioned test on wintel. Can anybody advice on how to
tune WL 8.1 on solaris 9 so as to make it perform best ? Is there any special
trick ?
thank u very much for any advice in advance
dso

"Arjan Kramer" <[email protected]> wrote:
>
Hi dso,
I'm running the same two configs and run into the same performance issues
as you do. Please let me know if you any response on this!
Regards,
Arjan Kramer
"dso" <[email protected]> wrote:
Hi All
I have setup my app to run on WL8.1 + solaris 9 env. JVM was configured
to use
3G RAM and there are still abundant RAM on the HW server. I tried out
a use case,
it took long time to get response (~ 2 minutes). But the CPU utilization
has been
always lower 20%. I have tried out the same test case on a wintel server
with
500 RAM allocated to JVM, the response time is much quicker (less than
30 sec).
I did the same on solaris 8 with 3G RAM and had used alternate threads
library
mode (changing LD_LIBRARY_PATH to include /usr/lib/lwp) which is the
default mode
adopted by solaris 9. The same use case responded much quicker and comparable
to abovementioned test on wintel. Can anybody advice on how to tuneWL
8.1 on
solaris 9 so as to make it perform best ? Is there any special trick
thank u very much for any advice in advance
dso
There could be many factors that add to performance degradation, database, OS,
Network, app config etc, so without knowing too much its difficult to tell.
Can you please supply the startup JAVA options used to set the heap etc. Having
larger heao sizes is not always the best approach of building HA applications...the
bigger they are, the bigger they fall. I'd suggest using many but smaller instances.
Provide the heap info from NT also.
BTW, when weblogic starts, can you tell me how much memory is being used in the
console...ie the footprint of weblogic + your application.
Many Thanks

IWS 6.0 100% CPU utilization hanging- very urgent

Hi,
We are using Iplanet Web server 6.0 on windows-2000 SP2.The problem we are facing is after 10 concurrent users have logged in the CPU utilization shoots up to 100% and we have to reboot the systesm
Our billing Application is affected very much due to this.
Can anybody throw some light on this?
Thanks in advance.

Hi,
Are you using any plugin with iWS. Please let me know your config file. Mean while please check tunning parameters of solaris for Performance bench mark.
http://docs.iplanet.com/docs/manuals/enterprise/50/tuning/perf6.htm#17580
Regards,
Dakshin.

How to check CPU % Utilization with SNMP

We are using Ipswitch's What's Up Professional 2006 to monitor devices on our network. Their default snmp graphing utilities use the HOST RESOURCES MIB to collect and graph system performance statistics over time. I have them running perfectly on all my Windows servers, but for my Solaris systems I am only able to retrieve Memory, Disk, and Network Interface statistics. The CPU monitor is unable to retrieve data from our Solaris systems. A typical system is running Solaris 10 with the default agent, NET-SNMP version 5.0.x. Does anyone know if the default agent supports checking CPU usage in the HOST RESOURCES MIB? If not can anyone point me to a different MIB & instance that will return % CPU utilization of a Solaris 10 server? Any input would be greatly appreciated. I've searched everywhere for the solution but I am unable to find any straight-forward answers, hopefully someone can help.

Hi
You can do it via T Code ST02.I am sending you a link hope it help you
<a href="http://help.sap.com/saphelp_erp2004/helpdata/en/02/96263e538111d1891b0000e8322f96/content.htm">Check this</a>
Rewards point if helpful
Thanks
Pankaj Kumar

NSAPI plugin has high CPU utilization on Sunone web server 6.0 SP5

Hi,
I am running WL 6.1 SP3 with pluginProxy SP03 on Sunone web server 6.0SP5 on Solaris.
Seeing very high CPU utilization with 3 threads running wl_proxy (about 33% each).
Any latest NSAPI pluginproxy patch I can use to fix this?
Walter

I'm having the same problems as all the above posts. I run a colloborative tool which uses IPlanet as a directory server and I receive the Event ID:25 Source: WebServer 6.0 error as well as Event ID:0 Source:https-admnserv6.0 which gives "the local computer may not have the necessary registry information or message DLL files to display messages from a remote computer". I have 3 servers built and all exhibit the same errors.

DS6 instance reaches 50% CPU utilization after restart

Hi,
I am running DS6 on Solaris 10. I noticed that after every orderly restart slapd process reaches 50% CPU utilization. This situation lasts for about 5 min. What do you think it is?
Thank you.

At Startup Directory Server does initialize some of its caches :
- ACI
- Roles
- Class Of Service.
- Groups (for the new isMemberOf feature).
Depending on your configuration these searches may take a little bit of time and cpu.
Regards,
Ludovic

High CPU utilization with JDesktopPane.OUTLINE_DRAG_MODE

Hello there,
since I updated from Java SDK 1.4.0 to 1.4.1_01 I recognized a problem with MDI Java applications using a JDesktopPane with JInternalFrames. When the drag mode of the internal frames is set to OUTLINE_DRAG_MODE, which should have a better performance than the LIVE_DRAG_MODE the cpu utilization goes nearly up to 100% an the drag of the frame is quite slow.
Does anybody else experience this problem?
(The problem exists in the application I develop and also in the IDE Netbeans, I use for development)
I am not sure if this is the right place for my problem, so if there is a better one to post it to, please tell me.
Thanks
R�diger

Hi,
I've also noticed this. It happens on Windows 2000 with 1.4.1, but not with version 1.4.0. Have you found a solution yet?
Martin

CUCM 8.6(2) - High CPU utilization - CCM process

Hello,
Process name ccm is peacking 91% of CPU utilization in one of Subscribers.
On cluster there are 1 Pub, 2 Subs, and 2 TFTP.
CCM service was already restart, and Subs reload, but problem remain.
We have checked and removed association with phones not registered, but problem remain.
Proc,PID,CPU,Status,SharedMem,Nice(Level)VmRSS,VMSize,VMData,ThreadCount,DataStackSize,PageFault
ccm,32687,90,SLEEPING,87380,0,426792,605276,468544,47,418345,744
VMware Installation:
        1 vCPU: Intel(R) Xeon(R) CPU E5-2643 0 @ 3.30GHz
        Disk 1: 80GB
        4096 Mbytes RAM
admin:show process name ccm
PID PPID   TID %CPU S COMMAND
32687     1 32687 0.0 S ccm
32687     1 19549 0.0 S ccm
32687     1 19570 0.0 S ccm
32687     1 19639 0.0 S ccm
32687     1 20057 0.0 S ccm
32687     1 20331 0.0 S ccm
32687     1 20332 0.0 S ccm
32687     1 20333 0.0 S ccm
32687     1 20354 0.0 S ccm
32687     1 20360 0.0 S ccm
32687     1 20361 0.0 S ccm
32687     1 23336 0.0 S ccm
32687     1 23337 0.0 S ccm
32687     1 23338 0.0 S ccm
32687     1 23339 1.0 S ccm
32687     1 23340 0.1 S ccm
32687     1 23341 0.2 S ccm
32687     1 23342 0.0 S ccm
32687     1 23349 0.0 S ccm
32687     1 23988 0.0 S ccm
32687     1 23989 21.6 S ccm
32687     1 23990 0.7 S ccm
32687     1 24013 0.0 S ccm
32687     1 24043 0.0 R ccm
32687     1 24044 0.0 S ccm
32687     1 24045 0.0 S ccm
32687     1 24046 0.0 S ccm
32687     1 24047 0.0 S ccm
32687     1 24048 0.7 R ccm
32687     1 24049 0.2 S ccm
32687     1 24050 0.1 R ccm
32687     1 24051 0.0 S ccm
32687     1 24052 0.0 S ccm
32687     1 24053 0.0 S ccm
32687     1 24054 0.0 S ccm
32687     1 24091 0.0 S ccm
32687     1 24092 0.0 S ccm
32687     1 24093 0.0 S ccm
32687     1 24181 0.0 S ccm
32687     1 24182 0.0 S ccm
32687     1 24192 0.0 S ccm
32687     1 24277 0.1 S ccm
32687     1 24278 0.0 S ccm
32687     1 24279 0.0 S ccm
32687     1 24280 0.0 S ccm
32687     1 24281 0.0 S ccm
32687     1 24282 0.0 S ccm
admin:
Are there any specific tests, or logs prior to isolate this issue?

Hi,
A few points:
> Use one of the recommended OVA templates
> Check if the spike is random or occurs at a specific time. If it occurs at a specific time it could be coinciding with something like DRS backup, CDR load, busy hours etc.
> Minimum set of logs needed for one such high CPU spike instance are:
Detailed ccm traces from all nodes
RisDC Perfmon logs
Proglogs
Also, capture the output of the following:
utils diagnose test
utils core active list
show process load
HTH
Manish

Can we control user request w.r.t Memory and CPU utilization in Oracle 10g

Dear All,
We are having Production with Oracle 10.2.0.4 (5 Node RAC, 32Gb RAM each) running on RHEL5.2 with 12000 Users. We have some schema say FIN, HRMS, SALES, REPORT and many dedicated users for those schemas. We need to control the user request against these schema with respect to Memory (or CPU utilization)
Suppose users using FIN schema can use Maximum 40 % of Total Memory, HRMS schema can use Max 20%, SALES can use Max 20% and REPORT can use Max 20%.
Is it possible to create any Service in Server side to handle this type of scenareo or any existing service which can be customised to fullfil this?
Please suggest me.......
Thanks,
Tusar
Edited by: gohappy on Jan 27, 2011 5:59 AM
Edited by: gohappy on Jan 27, 2011 6:00 AM

JDBC 'applications' quite often don't use persistent connections, and often do not exit gracefully by calling 'exit' or 'disconnect'.
This means the session will continue to exist.
It also means, if you don't establish any form of connection pooling and/or dead connection detection, you can throw whatever amount of memory in the server, and you will continue to report
'Now problems is coming'. Apart from crippled English, the general lesson any DBA should know is how these 'applications' operate, and, contrary to some, you can never ever fight problems caused by applications,
by throwing memory and cpu at the problem.
When I read your text, I also assume the application is not using PrepareStatement calls and not using bindvariables, this is why your 'application' is burning the CPU.
Find those 'application developers', sue them, or better still : Beat them with a whip, and have them fix their 'crapplication'.
Paraphrasing William Jefferson Clinton: It's the application, stupid!
Sybrand Bakker
Senior Oracle DBA

Increase in CPU Utilization after migration from APEX 3.1.2 to APEX 3.2

Has any noticed any increase in CPU Utilization after migrating from APEX 3.1.2 to APEX 3.2?
Thanks,
Mark

Hi Mark,
Take a look at some of the usage reports within APEX (sessions, page views etc) to get an overall feel for where the time is being consumed.
You'll also find it useful if you can run a statspack report (or AWR, ASH etc) during a busy period to be able to drill down into where that CPU is being spent.
There's no magic answers here unfortunately, you need to track it down to where the time is being spent before working back up to find out where best to tune it.
John.
Blog: http://jes.blogs.shellprompt.net
Work: http://www.apex-evangelists.com
Author of Pro Application Express: http://tinyurl.com/3gu7cd
REWARDS: Please remember to mark helpful or correct posts on the forum, not just for my answers but for everyone!

Which metadata table in the BODS repository has data about CPU utilization of every job that gets executed?

I have to create a performance dashboard where I need to make a graph showing CPU Utilization for individual workloads. If every job corresponds to a particular workload, is there any metadata table that would contain how much CPU Utilization corresponds to a given job?

If i am not wrong, AL_Statistics and ALVW_History will have details of memory utilization.

Custom report for TOP 10 CPU Utilization machines from any group in aggregation last seven days

I want to create a custom report that contain list of TOP CPU Utilization of machines form any group.This report is create on last 7 days CPU utilization of all machine from a group.
What should be query for this report.

Hi,
Please refer to the links below:
SQL Query for TOP 10 Average CPU
https://social.technet.microsoft.com/Forums/systemcenter/en-US/8d9a2d0d-8761-4d1f-b194-b24aa65172e1/sql-query-for-top-10-average-cpu?forum=operationsmanagerreporting
How to use Report Builder to create custom reports in SCOM 2007
http://www.systemcentercentral.com/how-to-use-report-builder-to-create-custom-reports-in-scom-2007/
Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact [email protected]

CPU utilization on Solaris

Similar Messages

Maybe you are looking for