如何获取 RAM 和 CPU 时间消耗？ Python Linux 上的应用程序

Question

I'm trying to get the value of the CPU consumption and the RAM consumption of my Python App on a Linux embedded environnement using Psutil and Memory_Profiler . The thing is I can't get it done, or more accurately I think I don't understand exactly what I'm getting as a result. I find Psutil quite unclear about the values returned by the methods so I don't know exactly what I should use in my case.

I've written this but I think this can't give me what I actually want:

f = open("mem_info.txt", "w") 
txt = "Memory RAM : " + str(process.memory_info().vms) + " bytes"
txt3 = "RAM details : " + str(virtual_memory())
txt2 = "Memory Percent : " + str(process.memory_percent(memtype="vms")) + " %"
txt4 = "CPU Time  : " + str(process.cpu_times()) + " s"
TXT = txt + "\n" + txt2 + "\n\n" + txt3 + "\n\n" + txt4 + "\n"
f.write(TXT)
f.close()

Output:

Memory RAM : 605085696 bytes
Memory Percent : 19.297597042513864 %

RAM details : svmem(total=3135549440, available=2531147776, percent=19.3, used=417918976, free=104349696, active=1263509504, inactive=1590898688, buffers=13201408, cached=2600079360, shared=17809408, slab=109674496)

CPU Time  : pcputimes(user=0.54, system=0.07, children_user=6.99, children_system=0.34, iowait=0.61) s

Could someone tell me if I should use an other method? Or If I used the right one, what result should I look at? Should I like sum the different results printed in CPU Time? But if I do so it's way to long for being the time of CPU used... PLease Help me: :(

Answer 1

https://psutil.readthedocs.io/en/latest/index.html?highlight=oneshot#psutil.Process.oneshot

Utility context manager which considerably speeds up the retrieval of multiple process information at the same time. Internally different process info (eg name(), ppid(), uids(), create_time(), …) may be fetched by using the same routine, but only one value is returned and the others are discarded. When using this context manager the internal routine is executed once (in the example below on name()) the value of interest is returned and the others are cached. The subsequent calls sharing the same internal routine will return the cached value. The cache is cleared when exiting the context manager block. The advice is to use this every time you retrieve more than one information about the process. If you're lucky, you'll get a hell of a speedup.

I ran this:

import psutil
import time

p = psutil.Process(pid=10500)
with p.oneshot():
    while p.status() == 'running':
        print(p.name())  # execute internal routine once collecting multiple info
        print(p.cpu_times())  # return cached value
        print(p.memory_info().vms)  # return cached value
        print(p.cpu_percent())  # return cached value
        print(p.create_time())  # return cached value
        print(p.ppid())  # return cached value
        print(p.status())  # return cached value
        time.sleep(1)

On a multi-threaded python.exe (I'm on Windows). It returns information about the process and seems to be accurate when comparing with the Resource Monitor. What is it that you don't understand about your output?

Also, regarding cpu_times() :

Return system CPU times as a named tuple. Every attribute represents the seconds the CPU has spent in the given mode. The attributes availability varies depending on the platform:

user: time spent by normal processes executing in user mode; on Linux this also includes guest time
system: time spent by processes executing in kernel mode
idle: time spent doing nothing

Answer 2

You can use the top command also, to get the most accurate stats

Run " top " directly in your shell to get interactive view like windows task manager

OR for command line to save output in txt file

top -n 1 -b > top-output.txt

you can cron this by saving it as a shell script,

Fairly extensive documentation, included in almost all Linux distributions

https://man7.org/linux/man-pages/man1/top.1.html