Sometimes, system administrators are asked to generate 'dummy' load on CPU to test the load vs performance. Here is a short example which shows how to generate load per CPU or per Core basis (actually threads):
# perl -e 'while (--$ARGV[0] and fork) {}; while () {}' 4
The last digit "4" in this case indicates how many CPUs or threads or cores you want to load by this method. This indeed is a good idea to test the CPU performance when some of the threaded/virtual CPUs are loaded.
When I gave the above command, in other terminal window for same server, I can observe something like this:
# mpstat 5
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 0 0 10 213 105 17 0 0 0 0 38 0 2 0 98
1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 100
2 0 0 0 2 0 2 0 0 0 0 0 0 0 0 100
3 0 0 0 1 0 0 0 0 0 0 0 0 0 0 100
4 0 0 0 2 0 2 0 0 0 0 5 0 0 0 100
5 0 0 3 4 0 6 0 0 0 0 0 0 0 0 100
6 0 0 1 13 6 11 0 0 0 0 0 0 0 0 100
7 0 0 4 3 0 4 0 0 0 0 10 0 0 0 100
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 21 7 139 227 118 18 9 0 1 0 53 85 1 0 14
1 4 0 7 9 0 16 0 4 0 0 4 1 1 0 98
2 10 0 0 8 0 5 5 0 1 0 1 85 0 0 15
3 0 1 34 7 0 11 0 1 0 0 3 0 0 0 100
4 6 0 0 6 0 2 4 0 0 0 4 84 0 0 16
5 7 0 0 2 0 1 0 1 1 0 1 0 0 0 100
6 10 0 4 12 5 3 5 0 0 0 1 85 0 0 15
7 0 0 21 16 0 31 0 0 0 0 10 0 0 0 100
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 0 0 30 207 106 0 8 0 0 0 0 100 0 0 0
1 0 0 0 9 0 15 0 1 0 0 23 0 1 0 99
2 0 0 0 6 0 0 5 0 0 0 0 100 0 0 0
3 0 0 3 8 0 13 0 1 0 0 12 0 0 0 100
4 0 0 0 6 0 0 5 0 0 0 0 100 0 0 0
5 0 0 0 2 0 0 0 0 0 0 0 0 0 0 100
6 0 0 1 10 3 2 7 0 0 0 0 100 0 0 0
7 0 0 0 1 0 0 0 0 0 0 0 0 0 0 100
^C
The 'top' command output in this case will look like following:
# top
last pid: 698; load avg: 1.61, 0.44, 0.19; up 25+19:31:24 10:59:01
34 processes: 29 sleeping, 5 on cpu
CPU states: 49.8% idle, 50.0% user, 0.2% kernel, 0.0% iowait, 0.0% swap
Memory: 8192M phys mem, 6680M free mem, 4099M total swap, 4099M free swap
PID USERNAME LWP PRI NICE SIZE RES STATE TIME CPU COMMAND
694 root 1 0 0 6368K 3256K cpu/0 0:35 12.50% perl
696 root 1 0 0 6360K 1672K cpu/7 0:35 12.45% perl
697 root 1 0 0 6360K 1672K cpu/1 0:35 12.45% perl
695 root 1 0 0 6360K 1800K cpu/6 0:35 12.14% perl
698 root 1 59 0 2944K 2104K cpu/3 0:00 0.03% top
679 root 1 59 0 10M 3576K sleep 0:00 0.00% sshd
119 root 24 59 0 4576K 4232K sleep 1:25 0.00% nscd
9 root 15 59 0 10M 9704K sleep 0:42 0.00% svc.configd
238 root 4 59 0 7888K 4664K sleep 0:29 0.00% inetd
7 root 12 59 0 13M 9904K sleep 0:34 0.00% svc.startd
302 root 16 59 0 15M 9400K sleep 0:27 0.00% fmd
202 root 1 59 0 1360K 1176K sleep 0:04 0.00% utmpd
183 daemon 1 59 0 2848K 2360K sleep 0:11 0.00% rpcbind
253 root 3 59 0 7464K 3160K sleep 0:08 0.00% automountd
1 root 1 59 0 2512K 1920K sleep 0:05 0.00% init
# perl -e 'while (--$ARGV[0] and fork) {}; while () {}' 4
The last digit "4" in this case indicates how many CPUs or threads or cores you want to load by this method. This indeed is a good idea to test the CPU performance when some of the threaded/virtual CPUs are loaded.
When I gave the above command, in other terminal window for same server, I can observe something like this:
# mpstat 5
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 0 0 10 213 105 17 0 0 0 0 38 0 2 0 98
1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 100
2 0 0 0 2 0 2 0 0 0 0 0 0 0 0 100
3 0 0 0 1 0 0 0 0 0 0 0 0 0 0 100
4 0 0 0 2 0 2 0 0 0 0 5 0 0 0 100
5 0 0 3 4 0 6 0 0 0 0 0 0 0 0 100
6 0 0 1 13 6 11 0 0 0 0 0 0 0 0 100
7 0 0 4 3 0 4 0 0 0 0 10 0 0 0 100
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 21 7 139 227 118 18 9 0 1 0 53 85 1 0 14
1 4 0 7 9 0 16 0 4 0 0 4 1 1 0 98
2 10 0 0 8 0 5 5 0 1 0 1 85 0 0 15
3 0 1 34 7 0 11 0 1 0 0 3 0 0 0 100
4 6 0 0 6 0 2 4 0 0 0 4 84 0 0 16
5 7 0 0 2 0 1 0 1 1 0 1 0 0 0 100
6 10 0 4 12 5 3 5 0 0 0 1 85 0 0 15
7 0 0 21 16 0 31 0 0 0 0 10 0 0 0 100
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 0 0 30 207 106 0 8 0 0 0 0 100 0 0 0
1 0 0 0 9 0 15 0 1 0 0 23 0 1 0 99
2 0 0 0 6 0 0 5 0 0 0 0 100 0 0 0
3 0 0 3 8 0 13 0 1 0 0 12 0 0 0 100
4 0 0 0 6 0 0 5 0 0 0 0 100 0 0 0
5 0 0 0 2 0 0 0 0 0 0 0 0 0 0 100
6 0 0 1 10 3 2 7 0 0 0 0 100 0 0 0
7 0 0 0 1 0 0 0 0 0 0 0 0 0 0 100
^C
The 'top' command output in this case will look like following:
# top
last pid: 698; load avg: 1.61, 0.44, 0.19; up 25+19:31:24 10:59:01
34 processes: 29 sleeping, 5 on cpu
CPU states: 49.8% idle, 50.0% user, 0.2% kernel, 0.0% iowait, 0.0% swap
Memory: 8192M phys mem, 6680M free mem, 4099M total swap, 4099M free swap
PID USERNAME LWP PRI NICE SIZE RES STATE TIME CPU COMMAND
694 root 1 0 0 6368K 3256K cpu/0 0:35 12.50% perl
696 root 1 0 0 6360K 1672K cpu/7 0:35 12.45% perl
697 root 1 0 0 6360K 1672K cpu/1 0:35 12.45% perl
695 root 1 0 0 6360K 1800K cpu/6 0:35 12.14% perl
698 root 1 59 0 2944K 2104K cpu/3 0:00 0.03% top
679 root 1 59 0 10M 3576K sleep 0:00 0.00% sshd
119 root 24 59 0 4576K 4232K sleep 1:25 0.00% nscd
9 root 15 59 0 10M 9704K sleep 0:42 0.00% svc.configd
238 root 4 59 0 7888K 4664K sleep 0:29 0.00% inetd
7 root 12 59 0 13M 9904K sleep 0:34 0.00% svc.startd
302 root 16 59 0 15M 9400K sleep 0:27 0.00% fmd
202 root 1 59 0 1360K 1176K sleep 0:04 0.00% utmpd
183 daemon 1 59 0 2848K 2360K sleep 0:11 0.00% rpcbind
253 root 3 59 0 7464K 3160K sleep 0:08 0.00% automountd
1 root 1 59 0 2512K 1920K sleep 0:05 0.00% init
No comments:
Post a Comment