Monitor performance

Hi:
I have a Sun Fire V440 box (2 CPU, 4G Ram, a 3510 disk storage).
I login from other machines (Sun Blade 2500), the response of the shell is very slow. However, the shell opened on the Sun Fire V440 box is reasonable. I suspect the network problem and thus change the network cables but problem still there.
I used RICHPse to monitor the disk. The performance of disk is slow. Below is the output from virtual_adrian.lite:
Adrian detected slow disk(s): Sat Mar 11 22:11:39 2006
disk r/s w/s Kr/s Kw/s wait actv svc_t %w %b delay
c1t1d0 264.2 38.0 10533.3 303.6 48.6 30.9 263.0 11 64 79502.2
c1t1d0s7 264.0 38.0 10533.3 303.6 48.6 30.9 263.2 11 64 79502.2
Adrian detected slow disk(s): Sat Mar 11 22:16:39 2006
disk r/s w/s Kr/s Kw/s wait actv svc_t %w %b delay
c1t1d0 318.1 17.8 14777.7 142.4 5.3 14.2 58.3 4 59 19579.3
c1t1d0s7 317.9 17.8 14777.7 142.4 5.3 14.2 58.3 4 59 19579.1
Adrian detected slow disk(s): Sat Mar 11 22:18:09 2006
disk r/s w/s Kr/s Kw/s wait actv svc_t %w %b delay
c1t1d0 62.9 17.3 3282.1 138.4 4.8 13.6 228.9 4 28 18352.8
c1t1d0s7 62.5 17.3 3281.9 138.4 4.8 13.6 230.0 4 28 18352.0
Adrian detected slow disk(s): Sat Mar 11 23:48:41 2006
disk r/s w/s Kr/s Kw/s wait actv svc_t %w %b delay
c1t1d0 6.1 42.3 46.9 11774.4 0.0 8.4 173.5 0 31 8403.3
c1t1d0s7 5.9 42.3 46.8 11774.2 0.0 8.4 173.9 0 31 8380.6
There is a lot of read and write and the disk response is slow. However, running top, I saw the CPU is idle.
The network card shall be fine as sometimes the link is okay.
I wonder if there is any method to find out the cause and fix it.
Thanks
Yoong

Seems obvious to the real me as well as the virtual me. Something is doing too much traffic to disk, so the system is unresponsive, idle CPU and network is what you should expect.
There isn't a standard way to tell which process is generating all the disk traffic, so unless its obvious by looking at the application code you could use a tool like Ortera Atlas (a.k.a. virtual_dave_fisk from www.ortera.com) to see how to tune the disk subsystem. I blogged on it a few times at http://perfcap.blogspot.com/2005/08/solving-storage-tuning-problems.html
Hope this helps
Adrian Cockcroft (now at eBay)

Similar Messages

How to monitor performance in Read Only data base

Hi,
We have a read only database that is using us to run reports against it.
1. How can i monitor performance in this read Only database ?
If i cant run statspack , how can i know what happened last hour for example
in the database ?
2. I use the glance utility to monitor the OS.
I see that the Disk Utilization is 100% and the swap is 80%.
Any suggestion what to do ?
Thanks

Hi,
Since the database is read only its cant be done:
clone SYS@pcrm > @spcreate.sql
... Installing Required Packages
create or replace package dbms_shared_pool is
ERROR at line 1:
ORA-00604: error occurred at recursive SQL level 1
ORA-16000: database open for read-only access
Thanks.

Monitoring & Performance Tuning

Hi BW Gurus,
Please let me know common Monitoring & Performance Tuning tasks/issues and how to resolve them.
Thanks in advance
Rajesh

hi Rajesh,
take a look sdn forum performance center
Business Intelligence Performance Tuning [original link is broken]
there are lots of docs
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/5ee957e5-0201-0010-9c83-fe14a43cd04a
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/cccad390-0201-0010-5093-fd9ec8157802
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/ce7fb368-0601-0010-64ba-fadc985a1f94
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/3f66ba90-0201-0010-ac8d-b61d8fd9abe9
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/4c0ab590-0201-0010-bd9a-8332d8b4f09c
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/1955ba90-0201-0010-d3aa-8b2a4ef6bbb2
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/c8c4d794-0501-0010-a693-918a17e663cc
and check oss note
557870 'FAQ BW Query Performance' and 567746 'Composite note BW 3.x performance Query and Web'
417307, 409641 etc (search with 'bw performance')
hope this helps.

Degraded Monitor Performance Post Snow Leopard Install

Hi.
I have noticed substantial degraded Monitor Performance since installing Snow Leopard. The Monitor is an NEC MultiSync 20WMGX2. The picture is "washed out" looking. Any ideas what the problem might be? The Monitor Settings are as followngs:
Brightness 50.0%
Contrast 43.0%
Sharpness 50.0%
Advanced DV Mode is "on"
DV Mode is set for Text
The Graphics Card is the standard Mac Mini Intel GMA950 with 64MB of shared system memory.
Thanks for any suggestions on how to get the performance back for this monitor.
Mike

Are you aware that the default display gamma has been switched from 1.8 to the default value of 2.2? If you want to build a 1.8 profile, then you still can. Some interpret the change as degradation.
8,961

Monitor performance/load of the portal application.

Hi,
is there any best tool to monitor performance/load of portal application.please help me out...?
Thanks
ramesh jillella
Edited by: Ramesh jillella on Mar 19, 2009 12:35 PM

I used YourKit on a project about a year ago. I've used a few other profilers and this one was the simplest to install, configure, and use in my opinion. I was doing basic CPU profiling of a WebLogic Portal application on my machine using YourKit within about a few hours of learning of the existence of YourKit. I used it to find performance bottlenecks and it did the job.
http://www.yourkit.com/home/index.jsp

How do i monitor performance in the system of a Java program in Windows XP?

I want to record the CPU usage and Memory usage on an hourly basis programatically, for an already running java class in Windows XP environment for the purpose of testing the Java class. The recording has to be done independent of the java class.By Cpu usage and Memory usage , i need to know how much of the computer's memory and CPU has been used by the java class.
I tried the HPROF tool available with J2SE 5.0, but it gives the details of each of the functions' usage in the program's allocated space in memory. moreover, it is not independent of the execution of the class.
The other option is to make use of the Performance tool available in Windows Xp->Administrative tools. This is a complex procedure, and i am not sure of how to proceed in this direction.

You can use the windows performance monitor (perfmon.exe / perfmon.msc). I don't know wther it is also included in the XP Home Edition. But this is the tool you are looking for. It can display the information in realtime but also can write to a protocoll.
Here's a shor description of what to do (i have a german windows so i'm more or less guessing the english terms)
- Start perfmon.msc
- In the tree on the you can see "systemonitor" (the realtime display) and "performance protocolls and warnings"
- open "performance protocalls and warnings", the first sub element should be something like"performance protocalls" or "performance indicator"
- open the context menu for it and create a new protocoll settings.
- add the information to it you are interested in and make other changes if you want
- start the new configuration
once you have the information you can display it it in the systemmonitor by selecting "display information from protocal" from the toolbar
It will take some time till you know what you are doing (and most of the information the performance monitor is useless to 99% of the people). But its a quite usfull tool once you know how to use it

ICM Monitoring & Performance Tuning

In out System >> DIALOG Avg Resp Time : 500 ms
But >> HTTP response time is huge around 3000 ms
How to monitor ICM Performance & Whih parameters help me to accelarate HTTP response.
Rgds

Hi,
You don't mention a release so taking a random one:
http://help.sap.com/saphelp_47x200/helpdata/en/28/c0133a62983c0ae10000000a114084/frameset.htm
on the left side you will see both parameters and monitoring for ICM.
parameters:
http://help.sap.com/saphelp_47x200/helpdata/en/61/f5183a3bef2669e10000000a114084/frameset.htm
monitor:
http://help.sap.com/saphelp_47x200/helpdata/en/ac/8f813a34872769e10000000a114084/frameset.htm
if you want to improve speed:
- check for non optimal coding ( if own development)
- check the buffers ( part of the parameters)
- check the timeout settings v
here is a demo profile for high load:
http://help.sap.com/saphelp_47x200/helpdata/en/ac/8f813a34872769e10000000a114084/frameset.htm
I believe there are also CCMS monitors available for ICM.
Hope this helps
Rgds
Nico

Looking for reliable programs to monitor performance

Any suggestions for managing the Mac-Pros performance?

Hi, not sure of Managing, but for Monitoring...
Open Activity Monitor, Show:>All Processes, sort on CPU%, see if anything using too much CPU% when this happens, click on Memory tab, do you have many Pageouts? Many tabs there to Monitor things.
Also, Open Console in Applications>Utilities & see if there are any clues or repeating messages.

Port monitoring performance issues

We have 3500Xl series switches and some are running port monitoring so we can gather data on what kind of traffic is flowing through our network. The question is if 1 ethernet port is monitoring all the other ethernet ports on it, what kind of performance issues could we expect, if any. Is there a way to show the performance stats of the switch as well. For example, see below.
interface FastEthernet0/47
description ethereal87 port monitor
no logging event link-status
port monitor FastEthernet0/2
port monitor FastEthernet0/1
port monitor FastEthernet0/4
port monitor FastEthernet0/3
port monitor FastEthernet0/6
port monitor FastEthernet0/5
port monitor FastEthernet0/8
port monitor FastEthernet0/10
port monitor FastEthernet0/9
port monitor FastEthernet0/12
port monitor FastEthernet0/11
port monitor FastEthernet0/14
port monitor FastEthernet0/13
port monitor FastEthernet0/16
port monitor FastEthernet0/15
port monitor FastEthernet0/20
port monitor FastEthernet0/19
port monitor FastEthernet0/22
port monitor FastEthernet0/21
port monitor FastEthernet0/24
port monitor FastEthernet0/23
port monitor FastEthernet0/26
port monitor FastEthernet0/25
port monitor FastEthernet0/28
port monitor FastEthernet0/27
port monitor FastEthernet0/30
port monitor FastEthernet0/29
port monitor FastEthernet0/32
port monitor FastEthernet0/31
port monitor FastEthernet0/34
port monitor FastEthernet0/33
port monitor FastEthernet0/36
port monitor FastEthernet0/35
port monitor FastEthernet0/38
port monitor FastEthernet0/37
port monitor FastEthernet0/40
port monitor FastEthernet0/39
port monitor FastEthernet0/42
port monitor FastEthernet0/41
port monitor FastEthernet0/44
port monitor FastEthernet0/43
port monitor FastEthernet0/46
port monitor FastEthernet0/45
switchport access vlan 87
no snmp trap link-status
spanning-tree portfast

Hi,
I noticed you are attempting this on a 3500XL switch.I'll attach a useful reference to some constraints with SPAN on XL
http://www.cisco.com/en/US/partner/products/hw/switches/ps708/products_tech_note09186a008015c612.shtml#topic1
These older switches use a blocking, shared memory architecture and shared buffers for groups of 4 ports. One performance condition that is possible in this type of scenario is intermittent packet loss within the group of ports when traffic is high. This condition becomes probable when utilisation of ports start exceeding about 70% of bandwidth simultaneously on all ports. The symptom is intermittent because of the nature of traffic is typically bursty
Example:
Monitor port is 48.
Servers are 45,46,47 -100MB
Let's say a backup is run on the servers and all servers start to hit 70% of available bandwidth. If port is monitoring, then this is = 3 x 70MB = 210MB which exceeds monitor port.
Monitor port would see overruns drops etc
http://www.cisco.com/en/US/partner/products/hw/switches/ps607/products_tech_note09186a0080125913.shtml#troubleshoot_interface_errors
but the server ports could intermittently start dropping traffic because of available buffer taken up by monitor port. There's no real way to determine at what rate this will happen but something to keep an eye out for. You may isolate the monitor port to it's own group of 4.
HTH
Eugene

System for monitor performance.

Hello Good Sirs and Madames
I am student in university and I study computer and enginerring
I have assign interesting
You are a consultant to a multi-national which wishes to reduce their IT budget without
sacrificing performance. As the technical person on the consulting team, you have
been assigned the task of assessing the performance and cost (TCO) of different
database systems on various operating systems and your report will be the basis
of the ultimate report submitted to the client.
The managing director of your client is particularly interested in reducing the licence
fees that his company pays to Oracle and wants the project to concentrate on
Open Source alternatives to Oracle.
The databases that shall be evaluated are:
Oracle
MySQL
Postgresql
Firebird
Derby/JavaDB
SQLite
These systems will be evaluated on Windows 2008 Server, Linux (CentOS), Solaris
and OpenBSD (insofar as this is possible).
The server will run an Apache web server and various standard workloads will be tested.
The project description is very long so I delete much. my question dear sirs and madames is about monitor software. my teacher say me that I first take "baseline" for all system and then install datbase and run test and monitor operating system performance. What are good way monitor operating system windows, linux and bsd?
I thank you kind Sirs and Madames for any answer.
Rajiv

Speaking to you as the person that wrote the Oracle curriculum at the University of Washington and taught it for ten years ... your description of your project is starting out with a D grade at best.
A database is not a spreadsheet with more horsepower. It is not a collection of bytes organized into rows and columns. You have missed the most important aspects of the assignment ... at least as I would grade it.
Here's what I would want to see you address.
1. Are there any governance and compliance issues? For example credit card numbers, tax identification numbers, proprietary financial data? If there are any replacement database must have encryption, security, secure backup, secure replication, auditing, and other capabilities. Oracle has them ... nothing else you listed does.
2. Is there an Service Level Agreement (SLA) that must be met? Can you achieve 99.99+% availability with all of these systems? Can you perform rolling patches? What is the support like when you have an outage? When you upgrade to new servers? To new storage? Replace failed or aged-out network components?
3. Is it scalable? Will it grow with the business requirements? Are you going to need to do something tomorrow you are not doing today? I just implemented a database with XML BINARY storage ... could I have done that in some inexpensive toy? What if I needed to store DICOM medical images? What if I needed to store non-scalar data types?
4. What is my ability to hire professionals in the marketplace that can deploy and manage the system long-term? Try to find an Oracle DBA ... easy in any country. Try to find a Firebird DBA? How many resume's are you going to be choosing from?
5. What is the stability of the company providing the product? Will they exist three years from now? Will they become part of IBM or Cisco or EMC or just be purchased for the pure joy of putting them out of business? For example where is the Ingres database today?
6. What is the overhead going to be migrating current database objects and code, as well as external application layer code to the new database product? Two years ago I participated in a project where the company was considering moving from Oracle to Teradata. The cost of rewriting the application was more than 300% the cost of the new hardware. They woke up and went to Exadata.
7. Do any of your current employees have production-level expertise with the new product? In the real world none of them will. No one in an Oracle shop has ever run production Derby. So assuming you could find a team that has are you going to fire your current employees? Or perhaps your plan is to train them and think they will enthusiastically want to learn something that will cut down their long-term employment prospects? Or send them to the university to learn it ... not that there are any classes or peer groups ... or user conferences, etc.
Real TCO, not kids-in-the-sandbox-TCO, requires that you have a price tag for every single aspect of the entire cost on the enterprise. Do you know the cost of hiring and training a new DBA?
If you have no database infrastructure you can look at TCO as a measure of which direction to travel. But if you have a mature organization the cost of change is almost always going to force you to stay exactly where you are with the possible exception of slowly moving to a new architecture on an incremental basis over many years.

New HighRes Monitor = performance issues

Hello,
So I recently got a new Monitor (LG 34UM95) wich is great, but my MacBook lags and is not as responsive as without the Monitor.
In Parallels and Inventor it's nearly unbearable, but it's even in Chrome or native OSX programs
Is the higher resolution really the problem, it's 3440x1440, I mean the Retina Display can easily be driven by a HD4000.
Here my Specs:
MacBook Pro 15" mid 2012
2,3 GHz i7
16GB 1600 Mhz
Nvidia 650M 512MB (I know 512MB ***** :/ )

I didn't mention I am running Leopard, but I have always run Leopard - I've had no experience in Tiger at all, so I'm hoping it's not related to the new O/S... there's a definite change in the way my macbook is running.

Account to monitor performance without sys credentials?

Hi folks!
I want to achieve the option for non sys users to get performance information out of the EMGC.
When they access the database and click on the performance link they will be asked for a username and password. I'm quite happy to create an db account that they can use (eg. perftestuser) but what priviledges does this account need in the database to get access to the relevant performance data that is displayed in the EMGC?
Any help is welcome.
cu
Andreas

Hi,
we have the same 'problem' - a read-only user for all.
We granted:
ROLES : OEM_MONITOR + SELECT_CATALOG_ROLE
SYSTEM PRIVILEGES: CREATE_SESSION
that's enough.
Best regards
Andreas

Monitoring Performance - HTTP 500

Hi all,
I have a system load testing and have been getting more than a 2:1 ratio of
Error 500's:
Error 500--Internal Server Error
From RFC 2068 Hypertext Transfer Protocol -- HTTP/1.1:
10.5.1 500 Internal Server Error
The server encountered an unexpected condition which prevented it from
fulfilling the request.
I am watching my default execute queue and it appears fine (a large
percentage of idle threads).. My default execute queue has 50 threads with
60% dedicated to socket connections (I have been playing around with this
number) - My test environment has 40 simultaneous threads making HTTP
requests with a 1 second delay between requests..
Other than viewing the results of my test, how can I detect when WLS is
returning an error and not processing a request (through JMX)?
The monitoring behavior I expected when I saw this error was a huge
utilization of threads in the execute queue - equal to 60% * 50 threads =
30, but I consistently have 35 - 40 threads idle.. Any ideas?
Furthermore I don't see any other database errors being generated in my log
files (and I have my servlet entry-point coded to throw a ServletException
when that occurs - so I should see that).
One last note, in my tests I have a request that does not invoke any
Servlets or JSP (strictly HTTP) and it is not generating this error..
Thanks in advance for your help!
Steve

Look in the weblogic.log file for a stack trace. That should tell you exactly
what is going on.
Mike
"Steven Haines" <[email protected]> wrote:
Hi all,
I have a system load testing and have been getting more than a 2:1 ratio
of
Error 500's:
Error 500--Internal Server Error
From RFC 2068 Hypertext Transfer Protocol -- HTTP/1.1:
10.5.1 500 Internal Server Error
The server encountered an unexpected condition which prevented
it from
fulfilling the request.
I am watching my default execute queue and it appears fine (a large
percentage of idle threads).. My default execute queue has 50 threads
with
60% dedicated to socket connections (I have been playing around with
this
number) - My test environment has 40 simultaneous threads making HTTP
requests with a 1 second delay between requests..
Other than viewing the results of my test, how can I detect when WLS
is
returning an error and not processing a request (through JMX)?
The monitoring behavior I expected when I saw this error was a huge
utilization of threads in the execute queue - equal to 60% * 50 threads
=
30, but I consistently have 35 - 40 threads idle.. Any ideas?
Furthermore I don't see any other database errors being generated in
my log
files (and I have my servlet entry-point coded to throw a ServletException
when that occurs - so I should see that).
One last note, in my tests I have a request that does not invoke any
Servlets or JSP (strictly HTTP) and it is not generating this error..
Thanks in advance for your help!
Steve

Need to monitor performance - best recommendation

We're going to be doing a dry-run of year-end processing and I was wondering if I could get some feedback on collecting some database performance data.
What should I be using etc...
I am running Oracle 9.2.0.4 on Solaris 9
Any input is appreciated.

You should change the format used for the snap_id column in the report. Open the source and change that - add some extra 999.
The same behaviour you can get on all large numbers which should be displayed in the report.
Here is a simple test:
SQL> column dummy format 999
SQL> select 99999 as dummy from dual;
DUMMY
HTH, Joze

How to monitor performance from OS

Hi,
DB : 10g R2
OS : AIX 5.3
I have more than 10 instances with 10 seprate databases on a Sigle Sever. Due to any database performance issue whole databases suffer due to single culprit DB . Is there any quick way from OS we see that which database is causing performance issue.
Quick help will be apprecaited.
Regards,

Hi,
Thanks for your reply.
where to start first? then next..... until conclusion?
I read awr report.I didn't understand the values.
Load Profile
<                      Per Second       Per Transaction Per Exec Per Call
*DB Time(s):           143.5               1.7                 0.19       0.13*
DB CPU(s):              5.3                 0.1                 0.01       0.00
Redo size:           29,998,607.4     355,473.0
Logical reads:        334,161.0         3,959.7
Block changes:       62,463.2          740.2
Physical reads:          5,678.0           67.3
Physical writes:          1,805.5          21.4
User calls:                  1,086.4         12.9
Parses:                         343.0          4.1
Hard parses:                  1.6              0.0
W/A MB processed:        17.0             0.2
Logons:                        8.4               0.1
Executes:                    742.0             8.8
Rollbacks:                     0.0                    0.0
Transactions:                84.4     For Ex:
What is maximun threshold value for DB time Per Second and per transacation?
I unable to understand these numbers?
Anyone help me.
Thanks & Regards,
VN

Monitor performance

Similar Messages

Maybe you are looking for