How to check process CPU and memory usage in Linux

Finding the exact process that is consuming CPU time or resident memory is the quickest path from “the server is slow” to a concrete fix on Linux systems, especially when load spikes, fan noise, or sudden OOM kills appear.

Process resource numbers come from the kernel’s procfs (/proc/), which tools like ps and top read to summarize per-process CPU and memory consumption. A ps listing provides a sortable snapshot, while top adds a sampled view that updates on an interval to expose short-lived spikes.

Percentages and sizes need the right interpretation: %CPU can vary based on sampling window and CPU count, %MEM is relative to total RAM, and a large VIRT value does not necessarily mean large physical memory use. When containers or cgroups are involved, limits can also make percentages look surprising compared to “host-wide” totals.

1. List the top CPU consumers with ps.

   $ ps -eo pid,comm,%cpu,%mem --sort=-%cpu | head -n 8
       PID COMMAND         %CPU %MEM
      3782 python3          2.8  0.4
         1 systemd          0.3  0.3
      2249 fwupd            0.2  1.0
       686 dbus-daemon      0.0  0.1
      3179 systemd          0.0  0.2
      3718 sshd             0.0  0.2
        11 kworker/u4:0-fl  0.0  0.0

2. List the top resident-memory consumers with ps.

   $ ps -eo pid,comm,rss,%mem --sort=-rss | head -n 8
       PID COMMAND           RSS %MEM
      2249 fwupd           41916  1.0
       364 multipathd      26240  0.6
       850 unattended-upgr 21888  0.5
      1593 packagekitd     19200  0.4
      3782 python3         18816  0.4
       311 systemd-journal 17368  0.4
         1 systemd         12660  0.3

   RSS is the resident set size shown in KiB by default.

3. Inspect the full command line and runtime for a heavy PID.

   $ ps -p 3782 -o pid,user,etime,pcpu,pmem,rss,args
       PID USER         ELAPSED %CPU %MEM   RSS COMMAND
      3782 root           00:01  2.7  0.4 18816 python3 -m http.server 9000 --bind 127.0.0.1

4. Capture a one-time sampled snapshot using top.

   $ top -b -n 1 | head -n 12
   top - 12:15:51 up 6 min,  2 users,  load average: 0.18, 0.10, 0.04
   Tasks: 123 total,   1 running, 121 sleeping,   0 stopped,   1 zombie
   %Cpu(s):  0.0 us,  4.2 sy,  0.0 ni, 95.8 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st 
   MiB Mem :   3901.5 total,   2923.8 free,    326.2 used,    809.2 buff/cache     
   MiB Swap:   2048.0 total,   2048.0 free,      0.0 used.   3575.3 avail Mem 
   
       PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
      3791 user      20   0   11752   4992   3072 R   9.1   0.1   0:00.03 top
         1 root      20   0   22416  12660   8564 S   0.0   0.3   0:01.43 systemd
         2 root      20   0       0      0      0 S   0.0   0.0   0:00.00 kthreadd
         3 root      20   0       0      0      0 S   0.0   0.0   0:00.00 pool_wo+
         4 root       0 -20       0      0      0 I   0.0   0.0   0:00.00 kworker+

   Use interactive top for live sorting: P sorts by %CPU and M sorts by %MEM.