Understanding TOP command and purpose

$top

top - 12:24:34 up 9 days, 21:58, 0 users, load average: 5.98, 5.32, 4.30 Tasks:
13 total, 1 running, 12 sleeping, 0 stopped, 0 zombie %Cpu(s): 5.5 us, 1.5 sy,
0.0 ni, 92.6 id, 0.0 wa, 0.0 hi, 0.5 si, 0.0 st KiB Mem: 12969522+total,
11112360+used, 18571628 free, 135900 buffers KiB Swap: 0 total, 0 used, 0 free.
49328208 cached Mem PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 40
root 20 0 1466540 912540 12568 S 7.7 0.7 67:03.03 bundle 43 root 20 0 1413152
860252 11116 S 6.7 0.7 65:41.24 bundle 

The load averages indicate the average number of processes waiting for CPU time over the specified time periods. Shows running processes and their status. Buffer is the amount of data used while it's being written or read. The numbers are in KiB's showing the RAM available on system us - user process sy - system process process ID (PID), user, priority (PR), virtual memory usage (VIRT), resident memory usage (RES), shared memory usage (SHR), CPU usage (%CPU), memory usage (%MEM), and the amount of time the process has been running (TIME+), COMMAND: command executed by each process. 

 

 kubectl top pods --containers --sort-by cpu
POD                                                NAME       CPU(cores)   MEMORY(bytes)   
dev-internal-default-fd58766c7-q2tpz   main       227m         109Mi           

dev-internal--default-fd58766c7-q2tpz   envoy      49m          47Mi            

dev-internal-default-fd58766c7-q2tpz   logstash   3m           573Mi  

clo-merchant-ita--itier--default-fd58766c7-q2tpz   filebeat   3m           68Mi            
-fd58766c7-q2tpz   filebeat   3m           68Mi            
      

#to check the no of threads in a process

$top -H 

Mem: 56073844K used, 8627892K free, 4648K shrd, 2704K buff, 26590424K cached
CPU:  17% usr   7% sys   0% nic  71% idle   0% io   0% irq   3% sirq
Load average: 5.73 9.04 9.89 7/7179 239
  PID  PPID USER     STAT   VSZ %VSZ CPU %CPU COMMAND
   99     1 root     S    12.0g  18%   1   1% node /var/groupon/node_modules/.bin/itier-server
  145     1 root     S    12.0g  18%   9   0% node /var/groupon/node_modules/.bin/itier-server
  131     1 root     S    12.0g  18%   0   0% node /var/groupon/node_modules/.bin/itier-server
  159     1 root     S    12.0g  18%   7   0% node /var/groupon/node_modules/.bin/itier-server
  161     1 root     S    12.0g  18%  11   0% node /var/groupon/node_modules/.bin/itier-server

$ kubectl describe PodMetrics appservice--default-74584b4f86-d2bzp
Name:         appservice--default-74584b4f86-d2bzp
Namespace:    appservice
Labels:       <none>
Annotations:  <none>
API Version:  metrics.k8s.io/v1beta1
Containers:
  Name:  main
  Usage:
    Cpu:     85935877n
    Memory:  201380Ki
  Name:      logstash
  Usage:
    Cpu:     915815n
    Memory:  627264Ki
  Name:      envoy
  Usage:
    Cpu:     27479576n
    Memory:  81324Ki
  Name:      filebeat
  Usage:
    Cpu:     1787023n
    Memory:  92932Ki
Kind:        PodMetrics
Metadata:
  Creation Timestamp:  2023-02-26T12:36:55Z
  Self Link:           /apis/metrics.k8s.io/v1beta1/namespaces/appservice/pods/appservice--default-74584b4f86-d2bzp
Timestamp:             2023-02-26T12:36:23Z
Window:                30s
Events:                <none>

There are 4 containers, main, logstash, envoy, filebeat. cpu is measured in nanoseconds referred as cpu nanocores and memory in KB