Performance Problems

[Heiner E. Lennackers]

Hi,

i have two identically equipped C3700 WS installed with the same OS-Load HP-UX 11.00 (installed from the same Ignite Server). But the users claim that one machine is significant slower than the other one. I checked very logfile and the hardware with stm, but cannot find a reason.
Are there some tools which i can use to get an objectiv view on the performance of the workstations? Maybe there are some problems (Hardware/Software), which cause the problem and i can find it, if i know in detail where the proformance get lost.

[Bharat Katkar]

HI,

1. Glance (this is the purchased product) is the best one to monitor System Performance.
2. #sar (System Activity Reporter) Use to monitor entire System Performance.
3. #iostat (Monitor Disk I/O)
4. #vmstat (Monitor's Virtual Memory)
6. #top (Monitors CPU, Processes, memory)
7. #netstat ( MOnitor network )

See man of all this command to learn more about them.

Also find the attached doc on system performance which may guide you further.

Hope that helps.
Regards,

[Steve Steel]

Hi

You got a good list there


Also look at

1)memory usage/swap see
Memory Usage - â What is using all of the memoryâ ?

by:eric.herberholz@hp.com

Last modified: July 23, 2004

Latest version available at external ftp site:

ftp://eh:spear9@hprc.external.hp.com/memory.htm


2)cpu and memory excesses

Top cpu usage . 5 lines

echo $_

user=$1

echo $user

echo " $(UNIX95= ps -e -o pcpu -o ruser -o sz -o vsz -opid -oppid -otty -o args|head -n1)"

UNIX95= ps -e -o pcpu -o ruser -o sz -o vsz -opid -oppid -otty -o args|grep -v %CPU| sort -nr|tail -n +2|head -n 20|grep " "$user



top memory usage 39 lines

1 #RSZ = Resident set size
2 #VSZ = Virtual set size
3 #
4 #RSZ is a subset of VSZ
5 #-the portion of the processes memory which is 'resident'
6 #in physical memory (ie not paged out).
7 #
8 #VSZ is the total size, including resident and non-resident pages.
9 #
10 #These values are totals, obtained by adding up the memory utilisation of the
11 #processes various memory segments (of which there can be lots).
12 #################################################################################
13 uname -a
14 echo -------------------------------------
15 /bin/rm /tmp/$PPID 2>/dev/null
16 if [ "$LOGNAME" = "root" ]
17 then
18 echo memory from kernel
19 name=$(file /stand/vmunix|grep 64)
20 if [ "$name" = "" ]
21 then
22 echo phys_mem_pages/D | adb /stand/vmunix /dev/kmem 2>/dev/null
23 else
24 echo phys_mem_pages/D | adb64 /stand/vmunix /dev/kmem 2>&1|grep -i page
25 fi
26 fi
27
28 echo " "
29 echo swap
30 echo -----------------
31 swapinfo -tm
32 echo " "
33 echo memory usage per process
34 echo "----------------------------------------------"
35 UNIX95= ps -e -o ruser,vsz,sz,pid,args > /tmp/$PPID
36 head -n 1 /tmp/$PPID
37 tail -n +2 /tmp/$PPID|
38 sort -rnk2
39 /bin/rm /tmp/$PPID 2>/dev/null


3)Do
swlist -l fileset -a state

On both machines and check for identical output


Steve

[Steve Lewis]

As well as the above, I would suggest that you do a few simple tests to verify the data rates from different bits of hardware.

1. Time how long it takes to FTP a 10Mb file to each server. This may point to a slow network connection (Check network interface speed and duplex using lanadmin -x ..., plus see what else is on that LAN)
2. Time how long it takes to copy a file from one directory to another.
3. Compare the backup times.
4. Check that the users are running the same applications and that one isn't being asked to do a lot more than the other.
5. If you have a big application that is installed on both servers, then set up tests to time how long it takes to execute the same task. For instance, a complex graphical rendering to test the graphics card etc.

[Steven E. Protter]

Good suggestion above, the port switch settings could be causing the whole problem.

Try this script set on both systems(attached) and look for other issues.

SEP

[Ted Buis]

Try nslookup on both machines to see if there are difference in name resolution. Also, the name switch service could be different.