py3nvml实现GPU相关信息读取的示例分析,很多新手对此不是很清楚,为了帮助大家解决这个难题,下面小编将为大家详细讲解,有这方面需求的人可以来学习下,希望你能有所收获。
在深度学习或者其他类型的GPU运算过程中,对于GPU信息的监测也是一个非常常用的功能。如果仅仅是使用系统级的GPU监测工具,就没办法非常细致的去跟踪每一步的显存和使用率的变化。如果是用profiler,又显得过于细致,而且环境配置、信息输出和筛选并不是很方便。此时就可以考虑使用py3nvml这样的工具,针对于GPU任务执行的过程进行细化的分析,有助于提升GPU的利用率和程序执行的性能。
技术背景
随着模型运算量的增长和硬件技术的发展,使用GPU来完成各种任务的计算已经渐渐成为算法实现的主流手段。而对于运行期间的一些GPU的占用,比如每一步的显存使用率等诸如此类的信息,就需要一些比较细致的GPU信息读取的工具,这里我们重点推荐使用py3nvml来对python代码运行的一个过程进行监控。
常规信息读取
一般大家比较常用的就是nvidia-smi
这个指令,来读取GPU的使用率和显存占用、驱动版本等信息:
$ nvidia-smiWed Jan 12 15:52:04 2022+-----------------------------------------------------------------------------+| NVIDIA-SMI 470.42.01 Driver Version: 470.42.01 CUDA Version: 11.4 ||-------------------------------+----------------------+----------------------+| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC || Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. || | | MIG M. ||===============================+======================+======================|| 0 Quadro RTX 4000 On | 00000000:03:00.0 On | N/A || 30% 39C P8 20W / 125W | 538MiB / 7979MiB | 16% Default || | | N/A |+-------------------------------+----------------------+----------------------+| 1 Quadro RTX 4000 On | 00000000:A6:00.0 Off | N/A || 30% 32C P8 7W / 125W | 6MiB / 7982MiB | 0% Default || | | N/A |+-------------------------------+----------------------+----------------------++-----------------------------------------------------------------------------+| Processes: || GPU GI CI PID Type Process name GPU Memory || ID ID Usage ||=============================================================================|| 0 N/A N/A 1643 G /usr/lib/xorg/Xorg 412MiB || 0 N/A N/A 2940 G /usr/bin/gnome-shell 76MiB || 0 N/A N/A 47102 G ...AAAAAAAAA= --shared-files 35MiB || 0 N/A N/A 172424 G ...AAAAAAAAA= --shared-files 11MiB || 1 N/A N/A 1643 G /usr/lib/xorg/Xorg 4MiB |+-----------------------------------------------------------------------------+
但是如果不使用profile仅仅使用nvidia-smi
这个指令的输出的话,是没有办法非常细致的分析程序运行过程中的变化的。这里顺便推荐一个比较精致的跟nvidia-smi
用法非常类似的小工具:gpustat。这个工具可以直接使用pip进行安装和管理:
$ python3 -m pip install gpustatCollecting gpustat Downloading gpustat-0.6.0.tar.gz (78 kB) |████████████████████████████████| 78 kB 686 kB/sRequirement already satisfied: six>=1.7 in /home/dechin/.local/lib/python3.8/site-packages (from gpustat) (1.16.0)Collecting nvidia-ml-py3>=7.352.0 Downloading nvidia-ml-py3-7.352.0.tar.gz (19 kB)Requirement already satisfied: psutil in /home/dechin/.local/lib/python3.8/site-packages (from gpustat) (5.8.0)Collecting blessings>=1.6 Downloading blessings-1.7-py3-none-any.whl (18 kB)Building wheels for collected packages: gpustat, nvidia-ml-py3 Building wheel for gpustat (setup.py) ... done Created wheel for gpustat: filename=gpustat-0.6.0-py3-none-any.whl size=12617 sha256=4158e741b609c7a1bc6db07d76224db51cd7656a6f2e146e0b81185ce4e960ba Stored in directory: /home/dechin/.cache/pip/wheels/0d/d9/80/b6cbcdc9946c7b50ce35441cc9e7d8c5a9d066469ba99bae44 Building wheel for nvidia-ml-py3 (setup.py) ... done Created wheel for nvidia-ml-py3: filename=nvidia_ml_py3-7.352.0-py3-none-any.whl size=19191 sha256=70cd8ffc92286944ad9f5dc4053709af76fc0e79928dc61b98a9819a719f1e31 Stored in directory: /home/dechin/.cache/pip/wheels/b9/b1/68/cb4feab29709d4155310d29a421389665dcab9eb3b679b527bSuccessfully built gpustat nvidia-ml-py3Installing collected packages: nvidia-ml-py3, blessings, gpustatSuccessfully installed blessings-1.7 gpustat-0.6.0 nvidia-ml-py3-7.352.0
使用的时候也是跟nvidia-smi非常类似的操作:
$ watch --color -n1 gpustat -cpu
返回结果如下所示:
Every 1.0s: gpustat -cpu ubuntu2004: Wed Jan 12 15:58:59 2022
ubuntu2004 Wed Jan 12 15:58:59 2022 470.42.01
[0] Quadro RTX 4000 | 39'C, 3 % | 537 / 7979 MB | root:Xorg/1643(412M) de
chin:gnome-shell/2940(75M) dechin:slack/47102(35M) dechin:chrome/172424(11M)
[1] Quadro RTX 4000 | 32'C, 0 % | 6 / 7982 MB | root:Xorg/1643(4M)
通过gpustat
返回的结果,包含了GPU的型号、使用率和显存使用大小和GPU当前的温度等常规信息。
py3nvml的安装与使用
接下来正式看下py3nvml的安装和使用方法,这是一个可以在python中实时查看和监测GPU信息的一个库,可以通过pip来安装和管理:
$ python3 -m pip install py3nvmlCollecting py3nvml Downloading py3nvml-0.2.7-py3-none-any.whl (55 kB) |████████████████████████████████| 55 kB 650 kB/sRequirement already satisfied: xmltodict in /home/dechin/anaconda3/lib/python3.8/site-packages (from py3nvml) (0.12.0)Installing collected packages: py3nvmlSuccessfully installed py3nvml-0.2.7
py3nvml绑定GPU卡
有一些框架为了性能的最大化,在初始化的时候就会默认去使用到整个资源池里面的所有GPU卡,比如如下使用Jax来演示的一个案例:
In [1]: import py3nvmlIn [2]: from jax import numpy as jnpIn [3]: x = jnp.ones(1000000000)In [4]: !nvidia-smiWed Jan 12 16:08:32 2022+-----------------------------------------------------------------------------+| NVIDIA-SMI 470.42.01 Driver Version: 470.42.01 CUDA Version: 11.4 ||-------------------------------+----------------------+----------------------+| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC || Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. || | | MIG M. ||===============================+======================+======================|| 0 Quadro RTX 4000 On | 00000000:03:00.0 On | N/A || 30% 41C P0 38W / 125W | 7245MiB / 7979MiB | 0% Default || | | N/A |+-------------------------------+----------------------+----------------------+| 1 Quadro RTX 4000 On | 00000000:A6:00.0 Off | N/A || 30% 35C P0 35W / 125W | 101MiB / 7982MiB | 0% Default || | | N/A |+-------------------------------+----------------------+----------------------++-----------------------------------------------------------------------------+| Processes: || GPU GI CI PID Type Process name GPU Memory || ID ID Usage ||=============================================================================|| 0 N/A N/A 1643 G /usr/lib/xorg/Xorg 412MiB || 0 N/A N/A 2940 G /usr/bin/gnome-shell 75MiB || 0 N/A N/A 47102 G ...AAAAAAAAA= --shared-files 35MiB || 0 N/A N/A 172424 G ...AAAAAAAAA= --shared-files 11MiB || 0 N/A N/A 812125 C /usr/local/bin/python 6705MiB || 1 N/A N/A 1643 G /usr/lib/xorg/Xorg 4MiB || 1 N/A N/A 812125 C /usr/local/bin/python 93MiB |+-----------------------------------------------------------------------------+
在这个案例中我们只是在显存中分配了一块空间用于存储一个向量,但是Jax在初始化之后,自动占据了本地的2张GPU卡。根据Jax官方提供的方法,我们可以使用如下的操作配置环境变量,使得Jax只能看到其中的1张卡,这样就不会扩张:
In [1]: import osIn [2]: os.environ["CUDA_VISIBLE_DEVICES"] = "1"In [3]: from jax import numpy as jnpIn [4]: x = jnp.ones(1000000000)In [5]: !nvidia-smiWed Jan 12 16:10:36 2022+-----------------------------------------------------------------------------+| NVIDIA-SMI 470.42.01 Driver Version: 470.42.01 CUDA Version: 11.4 ||-------------------------------+----------------------+----------------------+| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC || Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. || | | MIG M. ||===============================+======================+======================|| 0 Quadro RTX 4000 On | 00000000:03:00.0 On | N/A || 30% 40C P8 19W / 125W | 537MiB / 7979MiB | 0% Default || | | N/A |+-------------------------------+----------------------+----------------------+| 1 Quadro RTX 4000 On | 00000000:A6:00.0 Off | N/A || 30% 35C P0 35W / 125W | 7195MiB / 7982MiB | 0% Default || | | N/A |+-------------------------------+----------------------+----------------------++-----------------------------------------------------------------------------+| Processes: || GPU GI CI PID Type Process name GPU Memory || ID ID Usage ||=============================================================================|| 0 N/A N/A 1643 G /usr/lib/xorg/Xorg 412MiB || 0 N/A N/A 2940 G /usr/bin/gnome-shell 75MiB || 0 N/A N/A 47102 G ...AAAAAAAAA= --shared-files 35MiB || 0 N/A N/A 172424 G ...AAAAAAAAA= --shared-files 11MiB || 1 N/A N/A 1643 G /usr/lib/xorg/Xorg 4MiB || 1 N/A N/A 813030 C /usr/local/bin/python 7187MiB |+-----------------------------------------------------------------------------+
可以看到结果中已经是只使用了1张GPU卡,达到了我们的目的,但是这种通过配置环境变量来实现的功能还是着实不够pythonic,因此py3nvml中也提供了这样的功能,可以指定某一系列的GPU卡用于执行任务:
In [1]: import py3nvmlIn [2]: from jax import numpy as jnpIn [3]: py3nvml.grab_gpus(num_gpus=1,gpu_select=[1])Out[3]: 1In [4]: x = jnp.ones(1000000000)In [5]: !nvidia-smiWed Jan 12 16:12:37 2022+-----------------------------------------------------------------------------+| NVIDIA-SMI 470.42.01 Driver Version: 470.42.01 CUDA Version: 11.4 ||-------------------------------+----------------------+----------------------+| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC || Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. || | | MIG M. ||===============================+======================+======================|| 0 Quadro RTX 4000 On | 00000000:03:00.0 On | N/A || 30% 40C P8 20W / 125W | 537MiB / 7979MiB | 0% Default || | | N/A |+-------------------------------+----------------------+----------------------+| 1 Quadro RTX 4000 On | 00000000:A6:00.0 Off | N/A || 30% 36C P0 35W / 125W | 7195MiB / 7982MiB | 0% Default || | | N/A |+-------------------------------+----------------------+----------------------++-----------------------------------------------------------------------------+| Processes: || GPU GI CI PID Type Process name GPU Memory || ID ID Usage ||=============================================================================|| 0 N/A N/A 1643 G /usr/lib/xorg/Xorg 412MiB || 0 N/A N/A 2940 G /usr/bin/gnome-shell 75MiB || 0 N/A N/A 47102 G ...AAAAAAAAA= --shared-files 35MiB || 0 N/A N/A 172424 G ...AAAAAAAAA= --shared-files 11MiB || 1 N/A N/A 1643 G /usr/lib/xorg/Xorg 4MiB || 1 N/A N/A 814673 C /usr/local/bin/python 7187MiB |+-----------------------------------------------------------------------------+
可以看到结果中也是只使用了1张GPU卡,达到了跟上一步的操作一样的效果。
查看空闲GPU
对于环境中可用的GPU,py3nvml的判断标准就是在这个GPU上已经没有任何的进程,那么这个就是一张可用的GPU卡:
In [1]: import py3nvmlIn [2]: free_gpus = py3nvml.get_free_gpus()In [3]: free_gpusOut[3]: [True, True]
当然这里需要说明的是,系统应用在这里不会被识别,应该是会判断守护进程。
命令行信息获取
跟nvidia-smi
非常类似的,py3nvml也可以在命令行中通过调用py3smi
来使用。值得一提的是,如果需要用nvidia-smi
来实时的监测GPU的使用信息,往往是需要配合watch -n
来使用的,但是如果是py3smi
则不需要,直接用py3smi -l
就可以实现类似的功能。
$ py3smi -l 5Wed Jan 12 16:17:37 2022+-----------------------------------------------------------------------------+| NVIDIA-SMI Driver Version: 470.42.01 |+---------------------------------+---------------------+---------------------+| GPU Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |+=================================+=====================+=====================+| 0 30% 39C 8 19W / 125W | 537MiB / 7979MiB | 0% Default || 1 30% 33C 8 7W / 125W | 6MiB / 7982MiB | 0% Default |+---------------------------------+---------------------+---------------------++-----------------------------------------------------------------------------+| Processes: GPU Memory || GPU Owner PID Uptime Process Name Usage |+=============================================================================++-----------------------------------------------------------------------------+
可以看到略有区别的是,这里并不像nvidia-smi
列出来的进程那么多,应该是自动忽略了系统进程。
单独查看驱动版本和显卡型号
在py3nvml中把查看驱动和型号的功能单独列了出来:
In [1]: from py3nvml.py3nvml import *In [2]: nvmlInit()Out[2]: <CDLL 'libnvidia-ml.so.1', handle 560ad4d07a60 at 0x7fd13aa52340>In [3]: print("Driver Version: {}".format(nvmlSystemGetDriverVersion()))Driver Version: 470.42.01In [4]: deviceCount = nvmlDeviceGetCount() ...: for i in range(deviceCount): ...: handle = nvmlDeviceGetHandleByIndex(i) ...: print("Device {}: {}".format(i, nvmlDeviceGetName(handle))) ...:Device 0: Quadro RTX 4000Device 1: Quadro RTX 4000In [5]: nvmlShutdown()
这样也不需要我们自己再去逐个的筛选,从灵活性和可扩展性上来说还是比较方便的。
单独查看显存信息
这里同样的也是把显存的使用信息单独列了出来,不需要用户再去单独筛选这个信息,相对而言比较细致:
In [1]: from py3nvml.py3nvml import *In [2]: nvmlInit()Out[2]: <CDLL 'libnvidia-ml.so.1', handle 55ae42aadd90 at 0x7f39c700e040>In [3]: handle = nvmlDeviceGetHandleByIndex(0)In [4]: info = nvmlDeviceGetMemoryInfo(handle)In [5]: print("Total memory: {}MiB".format(info.total >> 20))Total memory: 7979MiBIn [6]: print("Free memory: {}MiB".format(info.free >> 20))Free memory: 7441MiBIn [7]: print("Used memory: {}MiB".format(info.used >> 20))Used memory: 537MiB
如果把这些代码插入到程序中,就可以获悉每一步所占用的显存的变化。
看完上述内容是否对您有帮助呢?如果还想对相关知识有进一步的了解或阅读更多相关文章,请关注编程网行业资讯频道,感谢您对编程网的支持。