DB2 Integrated Cluster Environment Deployment Guide

296 DB2 Integrated Cluster Environment Deployment Guide

Figure 5-94 Memory Visualizer

The Memory Visualizer (Figure 5-94) shows you (for each partition) a wide

variety of memory usage parameters, such as for the instance and databases.

You can have the values plotted to see how the trend is for a parameter.

Memory Visualizer is a graphical representation of db2mtrk data. db2mtrk data is

reported in the snapshot monitor stream in the Memory Heap section. In the

Memory Visualizer you also see to what extend your heaps are used, which

gives you the ability to detect the big allocated heaps.

5.2.2 Linux tools

All linux distributions come with a wide range of monitoring tools. We present

some of the most common tools delivered with Linux, and also some free,

available tools.

Chapter 5. Operations and administration 297

Vmstat

Vmstat reports virtual memory statistics, which include, in the given order, the

following information, shown in Example 5-7:

 r: Number of processes waiting for run time

 b: Number of processes in uninterruptible sleep

 swpd: Amount of virtual memory used

 free: Amount of idle memory

 buff: Amount of memory used as buffers

 cache: Amount of memory used as cache

 si: Memory swapped in from disk

 so: Memory swapped to disk

 bi: Blocks received

 bo: Blocks sent

 in: Interrupts (#)

 cs: Context switches (#)

 us: User time (%)

 sy: System time (%)

 id: Idle time (%)

 wa: I/O-waits (%)

You can run vmstat with a counter and a delay, for example, if you want vmstats

to collect the statistics four times every two seconds you use the following

command:

vmstat 2 4

The output of that command is shown in Example 5-7.

Example 5-7 vmstat

[root@DB2RB01 root]# vmstat 2 4

procs memory swap io system cpu

r b swpd free buff cache si so bi bo in cs us sy id wa

3 0 0 1098400 199052 2393244 0 0 4 36 0 61 39 1 58 2

3 0 0 1097660 199052 2393244 0 0 0 0 202 677 51 1 48 0

3 0 0 1096556 199052 2393244 0 0 0 60 136 522 51 3 45 1

3 0 0 1096556 199052 2393244 0 0 0 0 433 1337 50 1 49 0

This command is ideal for monitoring the CPU usage (us, sy, id, wa) and the

paging rate, which can be found under the memory in (si) and memory out (so)

columns. Other important columns are the amount of allocated virtual storage

(swpd) and free virtual storage (free).

This command is useful for determining if something is suspended or just taking

a long time. Additional information about vmstat can be found in the man pages

of vmstat (man vmstat).

298 DB2 Integrated Cluster Environment Deployment Guide

iostat

Iostats is used to show the throughput on the devices and partitions, and also

reports the CPU usage. The program gives you the possibility to run it multiple

times in a specified interval. Iostat is not installed by default (SuSE, Red Hat);

you have to install the sysstat package to get the iostat command.

The syntax of the command is:

iostat [ options... ] [ <interval> [ <count> ] ]

An example output of iostat 5 5 is shown in Example 5-8.

Example 5-8 iostat

[root@DB2RB01 root]# iostat 5 5

Linux 2.4.21-15.ELsmp (DB2RB01.torolab.ibm.com) 23/07/04

avg-cpu: %user %nice %sys %iowait %idle

39.22 0.12 1.48 1.64 57.55

Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn

sda 1.80 9.60 74.57 2451176 19033094

sda1 0.00 0.00 0.00 238 70

sda2 1.80 9.60 74.57 2450314 19033024

sda3 0.00 0.00 0.00 176 0

sdb 0.00 0.00 0.00 538 464

sdb1 0.00 0.00 0.00 338 464

avg-cpu: %user %nice %sys %iowait %idle

50.70 0.00 0.00 0.40 48.90

Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn

sda 0.60 0.00 43.29 0 216

sda1 0.00 0.00 0.00 0 0

sda2 0.60 0.00 43.29 0 216

sda3 0.00 0.00 0.00 0 0

sdb 0.00 0.00 0.00 0 0

sdb1 0.00 0.00 0.00 0 0

avg-cpu: %user %nice %sys %iowait %idle

52.00 0.00 1.10 0.00 46.90

Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn

sda 0.00 0.00 0.00 0 0

sda1 0.00 0.00 0.00 0 0

sda2 0.00 0.00 0.00 0 0

sda3 0.00 0.00 0.00 0 0

sdb 0.00 0.00 0.00 0 0

sdb1 0.00 0.00 0.00 0 0