Login or Register to Ask a Question and Join Our Community


AIX

DB2 high CPU

Login to Discuss or Reply to this Discussion in Our Community

 
Thread Tools Search this Thread
Operating Systems AIX DB2 high CPU
# 1  
Old 08-16-2010
DB2 high CPU
Hello dear friends, I have Lpar AIX 6.1 and there is DB2 installed. I have 8 Virtual CPU's configured on my Lpar and when I run nmon the CPU waiting time is always big.. I will provide screenshot for better realizing.. my question is what may produce so big waiting time?? Thanks in advance!
DB2 high CPU-nmon1jpeg
# 2  
Old 08-16-2010
Maybe The wait process means that CPU is IDLE and waiting for other jobs?
# 3  
Old 08-16-2010
Have you checked what processes are running currently? There can be several different reasons for that. Do you have a disk bottleneck (iostat) or network? Does DB2 have some internal trouble like locks that can't be resolved etc.?

Maybe this is interessting for you too:

Demystifying I/O Wait:
http://unix.derkeiler.com/Mailing-Li...ing_IOWAIT.pdf

---------- Post updated at 12:21 PM ---------- Previous update was at 12:16 PM ----------

Just saw your assumption - you can easily check if this is just the wait from the kernel wait processes if you run vmstat -w 1 10 for example. If the I/O Wait column is zero, then it is the wait processes.
I am not that familiar with that nmon view of the CPUs but I think that wait processes are not displayed in there, just real I/O wait like in vmstat for example.
# 4  
Old 08-16-2010
Code:
vmstat -w 1 10 

System configuration: lcpu=16 mem=24576MB ent=4.00

 kthr   memory                 page                       faults                 cpu          
-------  ------------------------------------ ------------------ -----------------------
  r   b        avm        fre    re    pi    po    fr     sr    cy    in     sy    cs us sy id wa    pc    ec
  2   0    5235358     868064     0     0     0     0      0     0  2999  10933  6223  5  3 88  4  0.40  10.1
  0   0    5235359     868063     0     0     0     0      0     0  2629   9390  5547  4  3 89  4  0.34   8.6
  0   0    5235357     868065     0     0     0     0      0     0  2547  13373  5302  5  4 88  4  0.40  10.0
  0   0    5235357     868065     0     0     0     0      0     0  3038  11937  6316  6  3 88  4  0.43  10.7
  0   0    5235357     868064     0     0     0     0      0     0  2421   9710  5455  4  3 89  4  0.35   8.8
  0   0    5235358     868063     0     0     0     0      0     0  2604   9692  5381  4  3 89  4  0.33   8.3
  0   0    5235358     868063     0     0     0     0      0     0  3722  11819  7843  5  4 87  4  0.45  11.3
  1   0    5235356     868065     0     0     0     0      0     0  3055  12577  6214  5  3 88  3  0.43  10.6
  0   0    5235357     868064     0     0     0     0      0     0  3354  20234  6494  6  4 86  3  0.51  12.8
  0   0    5235357     868064     0     0     0     0      0     0  2444   8784  5198  4  3 89  4  0.32   8.1


and here is iostat and I guess the only problem is that disks are heavily loaded
Code:
Disks:        % tm_act     Kbps      tps    Kb_read   Kb_wrtn
hdisk3          99.8     17080.0     897.1      75544     95384
hdisk2          94.6     16653.1     905.1      76600     90056
hdisk4          20.2     880.5     152.6         16      8796
hdisk0           1.4      28.4       2.8          4       280
hdisk1           1.3      28.0       2.7          0       280

# 5  
Old 08-16-2010
Yes there is some traffic on hdisk2 and hdisk3 having a high busy time and lots of transactions while using some bandwith.

Else CPU and memory wise the box looks good so that "wwwww"s in nmon will be the kernel wait processes, nothing to worry.

If you experience any slow behaviour though, you might want to think if it there is any improvement of the disk layout possible, ie. splitting tablespaces/containers to more than 2 volumes. Maybe also db2expln shows that there can be more efficiency by implementing an index at the right place etc.
I guess you have some SAN storage behind this with a cache - if this is the case, you might want to check if those SAN disks are ok with that load or not.

A vmstat -v might be interessting too in that case.
# 6  
Old 08-16-2010
Yes these heavily loaded disks are located on the DS8000 storadge.. but I have no good understanding how to work with this storadge and therefore I cant check anything on that storadge Smilie
and here is vmstat -v result and i want to thank you for your help...
Code:
vmstat -v
              6291456 memory pages
              6091921 lruable pages
               733113 free pages
                    3 memory pools
               483310 pinned pages
                 80.0 maxpin percentage
                  3.0 minperm percentage
                 90.0 maxperm percentage
                  4.5 numperm percentage
               278340 file pages
                  0.0 compressed percentage
                    0 compressed pages
                  4.5 numclient percentage
                 90.0 maxclient percentage
               278340 client pages
                    0 remote pageouts scheduled
                    0 pending disk I/Os blocked with no pbuf
                    0 paging space I/Os blocked with no psbuf
                 2484 filesystem I/Os blocked with no fsbuf
                54484 client filesystem I/Os blocked with no fsbuf
                    0 external pager filesystem I/Os blocked with no fsbuf
CDR /home #



---------- Post updated at 06:37 AM ---------- Previous update was at 06:34 AM ----------

and we are running out of storage space and in September we are going to make some upgrade... We think like you that splitting tablespaces/containers will give us some performance....
# 7  
Old 08-16-2010
Code:
54484 client filesystem I/Os blocked with no fsbuf

This is not pretty and could be tried to be tuned.

You can try:
Code:
ioo -p -o j2_dynamicBufferPreallocation=32

Default is 16. This is dynamic so remount is needed. If the counter still increases rather quick, you can do additionally:

Code:
ioo -p -o j2_nBufferPerPagerDevice=1024

Default is 512; you can also start with 2048. For this one you will have to remount the filesystems.

It could be that you do not notice this in terms of performance when using the application on top.
Login or Register to Ask a Question

Previous Thread | Next Thread

10 More Discussions You Might Find Interesting

1. Red Hat

Kswapd high cpu utilization.

Hi, Please suggest how to troubleshoot, kswapd is utilizing high cpu? also wanted to know which parameters are important and needs to be added with sar command for monitoring the performance of Linux (Oracle 5.8 -64 bit Please guide me. I am facing the issue where server is getting... (0 Replies)
Discussion started by: manoj.solaris
0 Replies

2. Shell Programming and Scripting

High cpu usage

I have created one script and i have added it into cron to run after 10 mins. However I have noted that whenever that script runs, It causes CPU utilization of server to increase about 10-20 % I have rechecked script and there is no way i can make changes, Script contain only 2-3 commands. So... (4 Replies)
Discussion started by: Nakul_sh
4 Replies

3. Red Hat

CPU is high

Hi , We found CPU is high due to python process .Is this something that Oracle team should look on or Unix team has to work on it ?Could you please advise use of python process ? top - 12:03:03 up 43 days, 15:11, 5 users, load average: 1.53, 1.33, 1.23 Tasks: 126 total, 3 running, 123... (12 Replies)
Discussion started by: Maddy123
12 Replies

4. Red Hat

High CPU Utilization

Hi Experts, I need to understand few basic things regarding top command result from one of the node i have collected: Cpu0 : 4.6%us, 2.0%sy, 0.0%ni, 91.4%id, 1.3%wa, 0.3%hi, 0.3%si, 0.0%st Cpu1 : 0.0%us, 0.3%sy, 0.0%ni, 99.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Cpu2 : ... (5 Replies)
Discussion started by: mackjack87
5 Replies

5. Shell Programming and Scripting

what would a script include to find CPU's %system time high and user time high?

Hi , I am trying to :wall: my head while scripting ..I am really new to this stuff , never did it before :( . how to find cpu's system high time and user time high in a script?? thanks , help would be appreciated ! :) (9 Replies)
Discussion started by: sushwey
9 Replies

6. Web Development

High CPU with Moodle

Hello list, I am managing a Moodle site on Apache2 server with Ubuntu. There are 20-30 simultaneous users and the server get 100% CPU and it stop working. The server is a dual core 3GHz and 4GB o memory. I think thta it is enough powerful machine for a Moodle site. How could I configure... (0 Replies)
Discussion started by: jlazkano
0 Replies

7. Red Hat

apache high cpu load on high traffic

i have a Intel Quad Core Xeon X3440 (4 x 2.53GHz, 8MB Cache, Hyper Threaded) with 16gig and 1tb harddrive with a 1gb port and my apache is causing my cpu to go up to 100% on all four cores heres my http.config <IfModule prefork.c> StartServers 10 MinSpareServers 10 MaxSpareServers 15... (4 Replies)
Discussion started by: awww
4 Replies

8. UNIX for Advanced & Expert Users

high cpu utilization

good morning. just wanted to ask if there's a way to check what causes the high cpu utilization of a server for the past 2 months? My jffnms report resulted to high utilization for a specific server last month. is there a way to check via a command line? thanks (9 Replies)
Discussion started by: lhareigh890
9 Replies

9. AIX

High CPU utilization

Hi am facing high cpu utilization on my sybase server. I have P550 Number Of Processors: 4 Processor Clock Speed: 1656 MHz CPU Type: 64-bit Kernel Type: 32-bit LPAR Info: 1 65-D837E Memory Size: 7840 MB in topas it shows Name PID CPU% PgSp Owner dataserv 565264 ... (1 Reply)
Discussion started by: vjm
1 Replies

10. Solaris

High CPU Utilization

Good morning, I need some help figuring out what's eating up my cpu. My application can't get enough cpu to do its job. this is a sunfire V440 2CPU's at 1/593 GHZ with 8GB of memory. In the morning hours the box is at less than 3%. I can't figure out what else is using the CPU. We use foglight and... (2 Replies)
Discussion started by: bbouhaik
2 Replies
Login or Register to Ask a Question