Python - 记录内存使用情况

Question

Python - 记录内存使用情况

pythonslurm

20

在Python 3中，有没有一种方法可以在某个程序运行时记录内存（RAM）使用情况？

一些背景信息。我在使用Slurm运行HPC集群上的仿真时，在提交作业之前必须预留一些内存。我知道我的工作需要很多内存，但我不确定需要多少。因此，我想知道是否有一种简单的解决方案来记录随时间变化的内存使用情况。

- physicsGuy

3个回答

2

您可以使用subprocess模块。这里是bash命令free的示例输出。

$ free -m
             total       used       free     shared    buffers     cached
Mem:          7979       7678        300          0        109       4628
-/+ buffers/cache:       2941       5038
Swap:         2046        360       1686

Python程序 -

import subprocess
result = subprocess.check_output(['bash','-c', 'free -m'])
free_memory = result.split('\n')[1].split()[3]
# print free_memory
# 300

如果您想检查某个进程的内存使用情况或定期记录它，那么您可以使用 pmap 或其他适合您用例的实用工具，然后解析输出。

- Jay

谢谢你的回答，如果不是因为damienfrancois的回答，我就会接受它了。 - physicsGuy

0

`bash`脚本跟踪Python（或其他）进程的资源使用情况

这里有一些例子脚本，可能会很方便。这样做的好处是，您不需要修改Python代码。缺点是：您需要一个bash。当然，您也可以从文件中运行Python脚本。将Python代码放在bash脚本中仅用于演示目的。

脚本

#!/bash/bin
set -e

LOG_OUTPUT="output.log"
LOG_RESOURCES="resources.log"

PYTHON_CODE="
import numpy as np
from datetime import datetime
import time

arr = np.zeros(1000)

print('start', datetime.now(), flush=True)
for ii in range(5000000):  # work
    arr += 1
    
print('break start', datetime.now(), flush=True)
time.sleep(3)  # break
print('break end', datetime.now(), flush=True)

for ii in range(5000000):  # work
    arr += 1
print('end', datetime.now(), flush=True)
"

rm -f ${LOG_OUTPUT}
rm -f ${LOG_RESOURCES}

log_process () {
  while ps -p $1 > /dev/null
  do
    top -p $1 -b -d 1 -n 1 >> $LOG_RESOURCES
    printf "\n\n" >> $LOG_RESOURCES
    sleep 1.
  done
}

python -c "${PYTHON_CODE}" >> ${LOG_OUTPUT} 2>&1 &  # run python in background
PID_PYTHON=$!  # python process id
echo "PID_PYTHON=${PID_PYTHON}"
tail -f ${LOG_OUTPUT} &  # show output on terminal
log_process $PID_PYTHON  # log needed resources to file

终端输出

$ bash monitor_python_script.sh 
PID_PYTHON=6807
start 2023-05-04 10:08:56.510581
end 2023-05-04 10:09:01.770897
break start 2023-05-04 10:09:04.280212
break end 2023-05-04 10:09:07.283365
end 2023-05-04 10:09:14.875994

日志文件输出：`output.log`

start 2023-05-04 10:08:56.510581
break start 2023-05-04 10:09:04.280212
break end 2023-05-04 10:09:07.283365
end 2023-05-04 10:09:14.875994

日志文件资源：`resources.log`

top - 10:08:56 up  1:47,  0 users,  load average: 0.08, 0.10, 0.09
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s): 16.1 us, 10.4 sy,  0.0 ni, 73.5 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29084.9 free,   3086.8 used,   2072.7 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30686.5 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:00.38 python


top - 10:08:57 up  1:47,  0 users,  load average: 0.08, 0.10, 0.09
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s): 14.6 us,  0.5 sy,  0.0 ni, 84.9 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29083.8 free,   3087.9 used,   2072.7 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30685.5 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:01.53 python


top - 10:08:58 up  1:47,  0 users,  load average: 0.23, 0.13, 0.10
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s): 15.2 us,  0.0 sy,  0.0 ni, 84.4 id,  0.5 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29083.9 free,   3087.8 used,   2072.8 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30685.6 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:02.69 python


top - 10:09:00 up  1:47,  0 users,  load average: 0.23, 0.13, 0.10
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s): 14.3 us,  0.0 sy,  0.0 ni, 85.7 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29083.1 free,   3088.5 used,   2072.8 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30684.9 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:03.85 python


top - 10:09:01 up  1:47,  0 users,  load average: 0.23, 0.13, 0.10
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s): 14.3 us,  0.0 sy,  0.0 ni, 85.7 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29083.3 free,   3088.3 used,   2072.8 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30685.1 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:05.01 python


top - 10:09:02 up  1:47,  0 users,  load average: 0.23, 0.13, 0.10
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s):  7.1 us,  0.5 sy,  0.0 ni, 92.0 id,  0.0 wa,  0.0 hi,  0.0 si,  0.5 st
MiB Mem :  34244.4 total,  29096.4 free,   3075.2 used,   2072.8 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30698.1 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:06.17 python


top - 10:09:03 up  1:47,  0 users,  load average: 0.29, 0.14, 0.11
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s):  7.1 us,  0.0 sy,  0.0 ni, 92.9 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29096.4 free,   3075.2 used,   2072.8 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30698.1 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:07.32 python


top - 10:09:04 up  1:47,  0 users,  load average: 0.29, 0.14, 0.11
Tasks:   1 total,   0 running,   1 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.0 us,  0.0 sy,  0.0 ni,100.0 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29096.9 free,   3074.7 used,   2072.8 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30698.6 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 S   0.0   0.1   0:08.05 python


top - 10:09:05 up  1:47,  0 users,  load average: 0.29, 0.14, 0.11
Tasks:   1 total,   0 running,   1 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.5 us,  0.0 sy,  0.0 ni, 99.5 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29096.6 free,   3074.9 used,   2073.0 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30698.3 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 S   0.0   0.1   0:08.05 python


top - 10:09:07 up  1:47,  0 users,  load average: 0.29, 0.14, 0.11
Tasks:   1 total,   0 running,   1 sleeping,   0 stopped,   0 zombie
%Cpu(s):  1.4 us,  0.5 sy,  0.0 ni, 98.1 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29087.0 free,   3084.6 used,   2072.8 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30688.8 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 S   0.0   0.1   0:08.05 python


top - 10:09:08 up  1:47,  0 users,  load average: 0.29, 0.14, 0.11
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s):  7.6 us,  0.0 sy,  0.0 ni, 92.4 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29086.5 free,   3085.1 used,   2072.9 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30688.4 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:08.95 python


top - 10:09:09 up  1:47,  0 users,  load average: 0.35, 0.16, 0.11
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s):  8.1 us,  0.0 sy,  0.0 ni, 91.9 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29086.7 free,   3084.9 used,   2072.9 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30688.6 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:10.11 python


top - 10:09:10 up  1:47,  0 users,  load average: 0.35, 0.16, 0.11
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s):  8.0 us,  0.0 sy,  0.0 ni, 92.0 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29087.0 free,   3084.5 used,   2072.9 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30688.9 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:11.27 python


top - 10:09:11 up  1:47,  0 users,  load average: 0.35, 0.16, 0.11
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s):  8.1 us,  0.0 sy,  0.0 ni, 91.9 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29086.5 free,   3085.1 used,   2072.9 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30688.4 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:12.43 python


top - 10:09:12 up  1:47,  0 users,  load average: 0.35, 0.16, 0.11
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s):  7.1 us,  0.0 sy,  0.0 ni, 92.9 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29086.3 free,   3085.3 used,   2072.9 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30688.2 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:13.59 python


top - 10:09:13 up  1:47,  0 users,  load average: 0.40, 0.17, 0.12
Tasks:   1 total,   1 running,   0 sleeping,   0 stopped,   0 zombie
%Cpu(s):  8.1 us,  0.0 sy,  0.0 ni, 91.9 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
MiB Mem :  34244.4 total,  29086.2 free,   3085.3 used,   2072.9 buff/cache
MiB Swap:   1024.0 total,   1024.0 free,      0.0 used.  30688.1 avail Mem 

    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
   6807 user      20   0  532700  28592  14444 R 100.0   0.1   0:14.74 python

- Markus Dutschke

网页内容由stack overflow 提供, 点击上面的

可以查看英文原文，
原文链接

- damienfrancois · Accepted Answer

你可以使用memory_profiler包来实现。只需在函数上添加一个装饰器@profile，你将得到如下输出：

Line #    Mem usage  Increment   Line Contents
==============================================
 3                           @profile
 4      5.97 MB    0.00 MB   def my_func():
 5     13.61 MB    7.64 MB       a = [1] * (10 ** 6)
 6    166.20 MB  152.59 MB       b = [2] * (2 * 10 ** 7)
 7     13.61 MB -152.59 MB       del b
 8     13.61 MB    0.00 MB       return a

否则，最简单的方法是使用命令sacct -l -j <JobId>向 Slurm 请求（查找 MaxRSS 列）以便您可以为进一步的作业进行调整。

另外，在运行程序时，您可以使用top 命令来了解其内存消耗情况。查找 RES 列即可。

Python - 记录内存使用情况

bash脚本跟踪Python（或其他）进程的资源使用情况

脚本

终端输出

日志文件输出：output.log

日志文件资源：resources.log

`bash`脚本跟踪Python（或其他）进程的资源使用情况

日志文件输出：`output.log`

日志文件资源：`resources.log`