dummy

1 System Monitor

1.1 GUI

pass

1.2 Command Line

1.2.1 vmstat

reports information about processes, memory, paging, block IO, traps, disks and cpu activity.

1.2.2 htop

htop is an interactive, commnad-line based system-monitor for viewing the running process on PC.

sudo apt install htop

1.2.3 glances

Glances is a free (LGPL) cross-platform curses-based system monitoring tool which aims to present a maximum of information in a minimum of space.

This tool is written in Python and uses the psutil library to fetch the statistical values from key elements, like CPU, load average, memory, network, disks, file systems, pro‐cesses and so on.

sudo apt-get install glances

Usage:

case1:

10-255-20-131 (debian stretch/sid 64bit / Linux 4.4.0-157-generic)   Uptime: 2:40:36

CPU      49.4%  nice:     0.4%    LOAD    4-core     MEM     41.9%  active:    3.50G   SWAP      0.0%
user:    31.1%  irq:      0.0%    1 min:    2.03     total:  7.80G  inactive:  1.71G   total:   8.00G
system:  17.5%  iowait:   1.7%    5 min:    1.31     used:   3.27G  buffers:   81.0M   used:        0
idle:    48.9%  steal:    0.0%    15 min:   1.02     free:   4.53G  cached:    2.58G   free:    8.00G

Processes filter: python (press ENTER to edit)
TASKS 9 (33 thr), 0 run, 9 slp, 0 oth sorted automatically by cpu_percent, flat view

  CPU%  MEM%  VIRT   RES   PID USER        NI S    TIME+ IOR/s IOW/s Command
  89.6  30.6 19.9G 2.38G 20424 root         0 D  3:45.60     0     0 python
   1.4   1.9 1.48G  155M  2023 root         0 S 36:46.18     0     0 python
   1.0  25.9 19.8G 2.02G 22214 root         0 S  0:02.28     0     0 python
   0.0   0.2 80.5M 16.2M  1922 root         0 S  0:02.16     0     0 python
   0.0   1.9 1.42G  154M  1942 root         0 S  0:02.10     0     0 python
   0.0   1.9 1.44G  148M 20399 root         0 S  0:02.14     0     0 python
   0.0   0.1 53.5M 10.7M 20430 root         0 S  0:00.40     0     0 python
   0.0  25.9 19.8G 2.02G 22212 root         0 S  0:02.22     0     0 python
   0.0  25.9 19.8G 2.02G 22213 root         0 S  0:02.24     0     0 python
   0.0  25.9 19.8G 2.02G 22215 root         0 S  0:02.22     0     0 python

case2:

10-255-20-131 (debian stretch/sid 64bit / Linux 4.4.0-157-generic)   Uptime: 2:50:22

PER CPU   50.0%  50.0%  50.0%  50.0%    LOAD    4-core     MEM     40.7%  active:    3.40G   SWAP      0.0%
user:      2.0%   1.0%   1.0%   2.0%    1 min:    2.18     total:  7.80G  inactive:  1.71G   total:   8.00G
system:    1.0%   2.0%   4.0%   0.0%    5 min:    2.06     used:   3.17G  buffers:   81.8M   used:        0
idle:      3.0%   3.0%   1.0%   2.0%    15 min:   1.57     free:   4.63G  cached:    2.58G   free:    8.00G

DISK I/O     R/s    W/s   TASKS 94 (220 thr), 1 run, 93 slp, 0 oth sorted automatically by cpu_percent, flat view
loop0          0      0
loop1          0      0     CPU%  MEM%  VIRT   RES   PID USER        NI S    TIME+ IOR/s IOW/s Command
loop2          0      0     76.7   0.3  108M 24.4M 30137 dc2-user     0 R  0:00.37     0     0 /usr/bin/python3 /usr/bin/glances --time 5 --percpu --hide-kernel-threads --disable-network
loop3          0      0     76.1  30.6 19.9G 2.38G 20424 root         0 S 13:32.55     0     0 python run_tasks.py --hypes /tmp/tmpf7uyhjqd/hypes.json --viz_pid 20399 --viz_port 8140 --tmp_d
loop4          0      0     30.2   1.9 1.48G  155M  2023 root         0 S 36:55.69     0     0 /usr/bin/python /home/dc2-user/work/qrs/torchcv-release/test/flask_services/cauchy_services.py
loop5          0      0      0.0   0.1 37.1M 5.84M     1 root         0 S  0:39.44     0     0 /sbin/init
loop6          0      0      0.0   0.0 34.6M 3.46M   661 root         0 S  0:02.57     0     0 /lib/systemd/systemd-journald
loop7          0      0      0.0   0.0 92.6M 1.52M   705 root         0 S  0:00.00     0     0 /sbin/lvmetad -f
vda            0      0      0.0   0.0 43.3M 3.77M   725 root         0 S  0:00.54     0     0 /lib/systemd/systemd-udevd
vda1           0      0      0.0   0.0 98.0M 2.49M   976 systemd-t    0 S  0:00.30     0     0 /lib/systemd/systemd-timesyncd
                             0.0   0.2  333M 15.3M  1143 root         0 S  0:00.30     0     0 /usr/sbin/smbd -D
FILE SYS    Used  Total      0.0   0.1  325M 5.71M  1145 root         0 S  0:00.00     0     0 /usr/sbin/smbd -D
/ (vda1)   62.9G  77.5G      0.0   0.1  333M 6.60M  1175 root         0 S  0:00.30     0     0 /usr/sbin/smbd -D
                             0.0   0.0 15.7M  868K  1177 root         0 S  0:00.00     0     0 /sbin/dhclient -1 -v -pf /run/dhclient.ens3.pid -lf /var/lib/dhcp/dhclient.ens3.leases -I -df /
                             0.0   0.0 5.10M  148K  1278 root         0 S  0:00.42     0     0 /sbin/iscsid
                             0.0   0.0 5.59M 3.43M  1279 root       -10 S  0:02.30     0     0 /sbin/iscsid
                             0.0   0.0 4.29M 1.20M  1283 root         0 S  0:00.00     0     0 /usr/sbin/acpid
                             0.0   0.0 31.2M 2.93M  1285 root         0 S  0:04.63     0     0 /usr/sbin/cron -f
                             0.0   0.0  250M 3.23M  1287 syslog       0 S  0:01.00     0     0 /usr/sbin/rsyslogd -n
                             0.0   0.0 93.1M 1.35M  1289 root         0 S  0:00.00     0     0 /usr/bin/lxcfs /var/lib/lxcfs/
                             0.0   0.0 25.4M 2.13M  1291 root         0 S  0:00.00     0     0 /usr/sbin/atd -f

1.2.4 lm-sensors

detect each relevant device (processors, fans, etc) and prepare to measure the temperature.

sudo apt-get install lm-sensors

Usage:

case:

1.2.5 nvidia-smi

GPU memory, power usage, temperature, fan speed and etc.

Usage:

case1: display a list of GPUs connected to the system

 GPU 0: Tesla P4 (UUID: GPU-e9cdea1d-74a3-d355-9f98-42ddf83c1893)

case2: device monitoring

# gpu  mclk  pclk   pwr gtemp mtemp    sm   mem   enc   dec
# Idx   MHz   MHz     W     C     C     %     %     %     %
    0  2999  1531    33    73     -     0     0     0     0
    0  2999  1531    33    72     -     0     0     0     0
    0  2999  1518    34    72     -    79    79     0     0

case3: process monitoring

# gpu        pid  type    sm   mem   enc   dec    fb   command
# Idx          #   C/G     %     %     %     %    MB   name
    0       5913     C    72    68     0     0  3225   python
    0       5913     C    66    61     0     0  3225   python
    0       5913     C    71    64     0     0  3225   python

参考1

参考2

1.2.6 gpustat

: GPU info, fan, power, temperature, pid

pip install gpustat

Usage:

case1:

[0] Tesla P4     | 44'C,  ?? %,   0 %,   28 /  75 W |  1059 /  7606 MB | root:python/7468(1049M)

case2: json