{"id":21293,"date":"2025-04-19T04:26:48","date_gmt":"2025-04-18T20:26:48","guid":{"rendered":"https:\/\/www.wsisp.com\/helps\/21293.html"},"modified":"2025-04-19T04:26:48","modified_gmt":"2025-04-18T20:26:48","slug":"%e6%9c%8d%e5%8a%a1%e5%99%a8gpu%e9%9b%86%e7%be%a4%e7%a1%ac%e4%bb%b6%e8%b5%84%e6%ba%90%e7%9b%91%e6%8e%a7%e5%ae%8c%e6%95%b4%e9%83%a8%e7%bd%b2%e6%8c%87%e5%8d%97","status":"publish","type":"post","link":"https:\/\/www.wsisp.com\/helps\/21293.html","title":{"rendered":"\u670d\u52a1\u5668GPU\u96c6\u7fa4\u786c\u4ef6\u8d44\u6e90\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357"},"content":{"rendered":"<h2>Grafana &#043; Prometheus &#043; DCGM&#xff1a;GPU\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357<\/h2>\n<p>\u6982\u8ff0&#xff1a;<\/p>\n<p>NVIDIA DCGM (Data Center GPU Manager) \u63d0\u4f9b\u4e86\u4e00\u5957\u5f3a\u5927\u7684\u5de5\u5177\u6765\u76d1\u63a7\u548c\u7ba1\u7406\u6570\u636e\u4e2d\u5fc3\u7684GPU\u3002\u7ed3\u5408Prometheus\u91c7\u96c6\u6570\u636e\u548cGrafana\u8fdb\u884c\u53ef\u89c6\u5316&#xff0c;\u53ef\u4ee5\u6784\u5efa\u4e00\u4e2a\u5168\u9762\u7684GPU\u96c6\u7fa4\u76d1\u63a7\u7cfb\u7edf\u3002\u672c\u6307\u5357\u5c06\u4ece\u96f6\u5f00\u59cb&#xff0c;\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u90e8\u7f72\u8fd9\u5957\u76d1\u63a7\u65b9\u6848\u3002<\/p>\n<h3>\u524d\u7f6e\u6761\u4ef6&#xff1a;<\/h3>\n<ul>\n<li>\u5b89\u88c5\u4e86NVIDIA\u9a71\u52a8\u7684Linux\u670d\u52a1\u5668<\/li>\n<li>\u5177\u6709root\u6216sudo\u6743\u9650<\/li>\n<li>Docker\u548cDocker Compose (\u63a8\u8350\u4f7f\u7528\u5bb9\u5668\u5316\u90e8\u7f72)<\/li>\n<li>\u57fa\u672c\u7684Linux\u547d\u4ee4\u884c\u77e5\u8bc6<\/li>\n<\/ul>\n<h3>\u7b2c\u4e00\u6b65&#xff1a;\u5b89\u88c5NVIDIA\u9a71\u52a8\u548cDCGM<\/h3>\n<p>\u9996\u5148\u786e\u4fdd\u60a8\u7684\u7cfb\u7edf\u5df2\u6b63\u786e\u5b89\u88c5NVIDIA\u9a71\u52a8&#xff1a;<\/p>\n<p>nvidia-smi<\/p>\n<p>\u5982\u679c\u663e\u793aGPU\u4fe1\u606f&#xff0c;\u8bf4\u660e\u9a71\u52a8\u5df2\u5b89\u88c5\u3002\u63a5\u4e0b\u6765\u5b89\u88c5DCGM&#xff1a;<\/p>\n<p>\u5bf9\u4e8eUbuntu\/Debian\u7cfb\u7edf&#xff1a;<\/p>\n<p><span class=\"token comment\"># \u6dfb\u52a0NVIDIA\u4ed3\u5e93<\/span><br \/>\n<span class=\"token assign-left variable\">distribution<\/span><span class=\"token operator\">&#061;<\/span><span class=\"token variable\"><span class=\"token variable\">$(<\/span><span class=\"token builtin class-name\">.<\/span> \/etc\/os-release<span class=\"token punctuation\">;<\/span><span class=\"token builtin class-name\">echo<\/span> $ID$VERSION_ID <span class=\"token operator\">|<\/span> <span class=\"token function\">sed<\/span> <span class=\"token parameter variable\">-e<\/span> <span class=\"token string\">&#039;s\/\\\\.\/\/g&#039;<\/span><span class=\"token variable\">)<\/span><\/span><br \/>\n<span class=\"token function\">wget<\/span> https:\/\/developer.download.nvidia.com\/compute\/cuda\/repos\/<span class=\"token variable\">$distribution<\/span>\/x86_64\/cuda-keyring_1.0-1_all.deb<br \/>\n<span class=\"token function\">sudo<\/span> dpkg <span class=\"token parameter variable\">-i<\/span> cuda-keyring_1.0-1_all.deb<\/p>\n<p><span class=\"token comment\"># \u66f4\u65b0\u5e76\u5b89\u88c5DCGM<\/span><br \/>\n<span class=\"token function\">sudo<\/span> <span class=\"token function\">apt-get<\/span> update<br \/>\n<span class=\"token function\">sudo<\/span> <span class=\"token function\">apt-get<\/span> <span class=\"token function\">install<\/span> <span class=\"token parameter variable\">-y<\/span> datacenter-gpu-manager<\/p>\n<p>\u5bf9\u4e8eCentOS\/RHEL\u7cfb\u7edf&#xff1a;<\/p>\n<p><span class=\"token comment\"># \u6dfb\u52a0NVIDIA\u4ed3\u5e93<\/span><br \/>\n<span class=\"token assign-left variable\">distribution<\/span><span class=\"token operator\">&#061;<\/span><span class=\"token variable\"><span class=\"token variable\">$(<\/span><span class=\"token builtin class-name\">.<\/span> \/etc\/os-release<span class=\"token punctuation\">;<\/span><span class=\"token builtin class-name\">echo<\/span> $ID$VERSION_ID <span class=\"token operator\">|<\/span> <span class=\"token function\">sed<\/span> <span class=\"token parameter variable\">-e<\/span> <span class=\"token string\">&#039;s\/\\\\.\/\/g&#039;<\/span><span class=\"token variable\">)<\/span><\/span><br \/>\n<span class=\"token function\">sudo<\/span> dnf config-manager &#8211;add-repo https:\/\/developer.download.nvidia.com\/compute\/cuda\/repos\/<span class=\"token variable\">$distribution<\/span>\/x86_64\/cuda-<span class=\"token variable\">$distribution<\/span>.repo<\/p>\n<p><span class=\"token comment\"># \u5b89\u88c5DCGM<\/span><br \/>\n<span class=\"token function\">sudo<\/span> dnf <span class=\"token function\">install<\/span> <span class=\"token parameter variable\">-y<\/span> datacenter-gpu-manager<\/p>\n<p>\u542f\u52a8DCGM\u670d\u52a1&#xff1a;<\/p>\n<p><span class=\"token function\">sudo<\/span> systemctl start nvidia-dcgm<br \/>\n<span class=\"token function\">sudo<\/span> systemctl <span class=\"token builtin class-name\">enable<\/span> nvidia-dcgm<\/p>\n<p>\u9a8c\u8bc1DCGM\u5b89\u88c5&#xff1a;<\/p>\n<p>dcgmi discovery <span class=\"token parameter variable\">-l<\/span><\/p>\n<h3>\u7b2c\u4e8c\u6b65&#xff1a;\u90e8\u7f72DCGM Exporter<\/h3>\n<p>DCGM Exporter\u662f\u8fde\u63a5DCGM\u548cPrometheus\u7684\u6865\u6881&#xff0c;\u5b83\u5c06DCGM\u6536\u96c6\u7684GPU\u6307\u6807\u8f6c\u6362\u4e3aPrometheus\u53ef\u4ee5\u6293\u53d6\u7684\u683c\u5f0f\u3002<\/p>\n<p>\u4f7f\u7528Docker\u90e8\u7f72DCGM Exporter&#xff08;\u63a8\u8350\u65b9\u5f0f&#xff09;&#xff1a;<\/p>\n<p><span class=\"token function\">docker<\/span> run <span class=\"token parameter variable\">-d<\/span> <span class=\"token parameter variable\">&#8211;gpus<\/span> all <span class=\"token parameter variable\">&#8211;restart<\/span> always <span class=\"token punctuation\">\\\\<\/span><br \/>\n  <span class=\"token parameter variable\">-p<\/span> <span class=\"token number\">9400<\/span>:9400 <span class=\"token punctuation\">\\\\<\/span><br \/>\n  <span class=\"token parameter variable\">&#8211;name<\/span> dcgm-exporter <span class=\"token punctuation\">\\\\<\/span><br \/>\n  nvcr.io\/nvidia\/k8s\/dcgm-exporter:2.4.6-2.6.10<\/p>\n<p>\u9a8c\u8bc1DCGM Exporter\u662f\u5426\u6b63\u5e38\u5de5\u4f5c&#xff1a;<\/p>\n<p><span class=\"token function\">curl<\/span> localhost:9400\/metrics<\/p>\n<p>\u60a8\u5e94\u8be5\u80fd\u770b\u5230\u5927\u91cf\u7684GPU\u6307\u6807\u6570\u636e\u3002<\/p>\n<h3>\u7b2c\u4e09\u6b65&#xff1a;\u5b89\u88c5\u548c\u914d\u7f6ePrometheus<\/h3>\n<p>\u521b\u5efaPrometheus\u914d\u7f6e\u6587\u4ef6&#xff1a;<\/p>\n<p><span class=\"token function\">mkdir<\/span> <span class=\"token parameter variable\">-p<\/span> \/opt\/prometheus<\/p>\n<p>\u521b\u5efa\u914d\u7f6e\u6587\u4ef6\/opt\/prometheus\/prometheus.yml&#xff1a;<\/p>\n<p><span class=\"token key atrule\">global<\/span><span class=\"token punctuation\">:<\/span><br \/>\n  <span class=\"token key atrule\">scrape_interval<\/span><span class=\"token punctuation\">:<\/span> 15s<br \/>\n  <span class=\"token key atrule\">evaluation_interval<\/span><span class=\"token punctuation\">:<\/span> 15s<\/p>\n<p><span class=\"token key atrule\">scrape_configs<\/span><span class=\"token punctuation\">:<\/span><br \/>\n  <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token key atrule\">job_name<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token string\">&#039;dcgm-exporter&#039;<\/span><br \/>\n    <span class=\"token key atrule\">static_configs<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token key atrule\">targets<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;localhost:9400&#039;<\/span><span class=\"token punctuation\">]<\/span><\/p>\n<p>  <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token key atrule\">job_name<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token string\">&#039;prometheus&#039;<\/span><br \/>\n    <span class=\"token key atrule\">static_configs<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token key atrule\">targets<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;localhost:9090&#039;<\/span><span class=\"token punctuation\">]<\/span><\/p>\n<p>\u4f7f\u7528Docker\u90e8\u7f72Prometheus&#xff1a;<\/p>\n<p><span class=\"token function\">docker<\/span> run <span class=\"token parameter variable\">-d<\/span> <span class=\"token parameter variable\">&#8211;restart<\/span> always <span class=\"token punctuation\">\\\\<\/span><br \/>\n  <span class=\"token parameter variable\">-p<\/span> <span class=\"token number\">9090<\/span>:9090 <span class=\"token punctuation\">\\\\<\/span><br \/>\n  <span class=\"token parameter variable\">&#8211;name<\/span> prometheus <span class=\"token punctuation\">\\\\<\/span><br \/>\n  <span class=\"token parameter variable\">-v<\/span> \/opt\/prometheus\/prometheus.yml:\/etc\/prometheus\/prometheus.yml <span class=\"token punctuation\">\\\\<\/span><br \/>\n  prom\/prometheus<\/p>\n<p>\u9a8c\u8bc1Prometheus\u662f\u5426\u6b63\u5e38\u5de5\u4f5c&#xff1a;\u8bbf\u95eehttp:\/\/your-server-ip:9090&#xff0c;\u60a8\u5e94\u8be5\u80fd\u770b\u5230Prometheus\u7684Web\u754c\u9762\u3002<\/p>\n<p>\u68c0\u67e5\u76ee\u6807\u662f\u5426\u6b63\u5e38&#xff1a;\u5728Prometheus\u754c\u9762\u5bfc\u822a\u5230Status &gt; Targets&#xff0c;\u786e\u8ba4dcgm-exporter\u76ee\u6807\u72b6\u6001\u4e3aUP\u3002<\/p>\n<h3>\u7b2c\u56db\u6b65&#xff1a;\u5b89\u88c5\u548c\u914d\u7f6eGrafana<\/h3>\n<p>\u4f7f\u7528Docker\u90e8\u7f72Grafana&#xff1a;<\/p>\n<p><span class=\"token function\">docker<\/span> run <span class=\"token parameter variable\">-d<\/span> <span class=\"token parameter variable\">&#8211;restart<\/span> always <span class=\"token punctuation\">\\\\<\/span><br \/>\n  <span class=\"token parameter variable\">-p<\/span> <span class=\"token number\">3000<\/span>:3000 <span class=\"token punctuation\">\\\\<\/span><br \/>\n  <span class=\"token parameter variable\">&#8211;name<\/span> grafana <span class=\"token punctuation\">\\\\<\/span><br \/>\n  grafana\/grafana<\/p>\n<p>\u8bbf\u95eeGrafana&#xff1a;\u6253\u5f00\u6d4f\u89c8\u5668\u8bbf\u95eehttp:\/\/your-server-ip:3000&#xff0c;\u9ed8\u8ba4\u7528\u6237\u540d\u548c\u5bc6\u7801\u90fd\u662fadmin\u3002<\/p>\n<p>\u914d\u7f6ePrometheus\u6570\u636e\u6e90&#xff1a;<\/p>\n<li>\u767b\u5f55\u540e&#xff0c;\u70b9\u51fb\u5de6\u4fa7\u83dc\u5355\u7684&#034;Configuration&#034;&#xff08;\u9f7f\u8f6e\u56fe\u6807&#xff09;<\/li>\n<li>\u9009\u62e9&#034;Data Sources&#034;<\/li>\n<li>\u70b9\u51fb&#034;Add data source&#034;<\/li>\n<li>\u9009\u62e9&#034;Prometheus&#034;<\/li>\n<li>\u5728URL\u5b57\u6bb5\u4e2d\u8f93\u5165http:\/\/prometheus:9090&#xff08;\u5982\u679cPrometheus\u548cGrafana\u4e0d\u5728\u540c\u4e00\u7f51\u7edc&#xff0c;\u4f7f\u7528\u670d\u52a1\u5668IP&#xff09;<\/li>\n<li>\u70b9\u51fb&#034;Save &amp; Test&#034;\u786e\u8ba4\u8fde\u63a5\u6210\u529f<\/li>\n<h3>\u7b2c\u4e94\u6b65&#xff1a;\u5bfc\u5165NVIDIA GPU\u76d1\u63a7\u9762\u677f<\/h3>\n<p>Grafana\u6709\u5f88\u591a\u73b0\u6210\u7684NVIDIA GPU\u76d1\u63a7\u9762\u677f&#xff0c;\u6211\u4eec\u53ef\u4ee5\u5bfc\u5165\u4f7f\u7528&#xff1a;<\/p>\n<li>\u5728Grafana\u5de6\u4fa7\u83dc\u5355\u70b9\u51fb&#034;&#043;\u201c\u56fe\u6807&#xff0c;\u9009\u62e9&#034;Import\u201d<\/li>\n<li>\u8f93\u5165\u9762\u677fID 12239&#xff08;\u8fd9\u662f\u4e00\u4e2a\u4f18\u79c0\u7684NVIDIA GPU\u76d1\u63a7\u9762\u677f&#xff09;<\/li>\n<li>\u70b9\u51fb&#034;Load&#034;<\/li>\n<li>\u9009\u62e9\u4e4b\u524d\u521b\u5efa\u7684Prometheus\u6570\u636e\u6e90<\/li>\n<li>\u70b9\u51fb&#034;Import&#034;\u5b8c\u6210\u5bfc\u5165<\/li>\n<p>\u60a8\u4e5f\u53ef\u4ee5\u5c1d\u8bd5\u5176\u4ed6\u9762\u677fID&#xff0c;\u598214574\u621615117&#xff0c;\u5b83\u4eec\u63d0\u4f9b\u4e0d\u540c\u98ce\u683c\u7684GPU\u76d1\u63a7\u89c6\u56fe\u3002<\/p>\n<h3>\u76d1\u63a7\u7684GPU\u6307\u6807\u8bf4\u660e&#xff1a;<\/h3>\n<p>\u5bfc\u5165\u7684\u9762\u677f\u5c06\u663e\u793a\u4ee5\u4e0b\u5173\u952eGPU\u6307\u6807&#xff1a;<\/p>\n<ul>\n<li>GPU\u5229\u7528\u7387&#xff1a;\u663e\u793aGPU\u8ba1\u7b97\u6838\u5fc3\u7684\u4f7f\u7528\u767e\u5206\u6bd4<\/li>\n<li>GPU\u5185\u5b58\u4f7f\u7528&#xff1a;\u663e\u793aGPU\u663e\u5b58\u7684\u4f7f\u7528\u60c5\u51b5<\/li>\n<li>\u6e29\u5ea6&#xff1a;GPU\u6838\u5fc3\u6e29\u5ea6<\/li>\n<li>\u529f\u8017&#xff1a;GPU\u7684\u529f\u8017\u6d88\u8017<\/li>\n<li>PCIe\u5e26\u5bbd&#xff1a;\u901a\u8fc7PCIe\u603b\u7ebf\u7684\u6570\u636e\u4f20\u8f93\u901f\u7387<\/li>\n<li>SM Clock&#xff1a;GPU\u6d41\u5904\u7406\u5668\u7684\u65f6\u949f\u9891\u7387<\/li>\n<li>Memory Clock&#xff1a;GPU\u5185\u5b58\u7684\u65f6\u949f\u9891\u7387<\/li>\n<li>Encoder\/Decoder\u5229\u7528\u7387&#xff1a;\u89c6\u9891\u7f16\u89e3\u7801\u5668\u7684\u4f7f\u7528\u60c5\u51b5<\/li>\n<\/ul>\n<h3>\u4f7f\u7528Docker Compose\u5b9e\u73b0\u4e00\u952e\u90e8\u7f72<\/h3>\n<p>\u4e3a\u4e86\u7b80\u5316\u90e8\u7f72&#xff0c;\u521b\u5efadocker-compose.yml\u6587\u4ef6&#xff1a;<\/p>\n<p><span class=\"token key atrule\">version<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token string\">&#039;3&#039;<\/span><\/p>\n<p><span class=\"token key atrule\">services<\/span><span class=\"token punctuation\">:<\/span><br \/>\n  <span class=\"token key atrule\">dcgm-exporter<\/span><span class=\"token punctuation\">:<\/span><br \/>\n    <span class=\"token key atrule\">image<\/span><span class=\"token punctuation\">:<\/span> nvcr.io\/nvidia\/k8s\/dcgm<span class=\"token punctuation\">&#8211;<\/span>exporter<span class=\"token punctuation\">:<\/span>2.4.6<span class=\"token punctuation\">&#8211;<\/span>2.6.10<br \/>\n    <span class=\"token key atrule\">restart<\/span><span class=\"token punctuation\">:<\/span> always<br \/>\n    <span class=\"token key atrule\">ports<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token string\">&#034;9400:9400&#034;<\/span><br \/>\n    <span class=\"token key atrule\">deploy<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token key atrule\">resources<\/span><span class=\"token punctuation\">:<\/span><br \/>\n        <span class=\"token key atrule\">reservations<\/span><span class=\"token punctuation\">:<\/span><br \/>\n          <span class=\"token key atrule\">devices<\/span><span class=\"token punctuation\">:<\/span><br \/>\n            <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token key atrule\">driver<\/span><span class=\"token punctuation\">:<\/span> nvidia<br \/>\n              <span class=\"token key atrule\">count<\/span><span class=\"token punctuation\">:<\/span> all<br \/>\n              <span class=\"token key atrule\">capabilities<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token punctuation\">[<\/span>gpu<span class=\"token punctuation\">]<\/span><\/p>\n<p>  <span class=\"token key atrule\">prometheus<\/span><span class=\"token punctuation\">:<\/span><br \/>\n    <span class=\"token key atrule\">image<\/span><span class=\"token punctuation\">:<\/span> prom\/prometheus<br \/>\n    <span class=\"token key atrule\">restart<\/span><span class=\"token punctuation\">:<\/span> always<br \/>\n    <span class=\"token key atrule\">ports<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token string\">&#034;9090:9090&#034;<\/span><br \/>\n    <span class=\"token key atrule\">volumes<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> .\/prometheus\/prometheus.yml<span class=\"token punctuation\">:<\/span>\/etc\/prometheus\/prometheus.yml<br \/>\n    <span class=\"token key atrule\">depends_on<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> dcgm<span class=\"token punctuation\">&#8211;<\/span>exporter<\/p>\n<p>  <span class=\"token key atrule\">grafana<\/span><span class=\"token punctuation\">:<\/span><br \/>\n    <span class=\"token key atrule\">image<\/span><span class=\"token punctuation\">:<\/span> grafana\/grafana<br \/>\n    <span class=\"token key atrule\">restart<\/span><span class=\"token punctuation\">:<\/span> always<br \/>\n    <span class=\"token key atrule\">ports<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token string\">&#034;3000:3000&#034;<\/span><br \/>\n    <span class=\"token key atrule\">volumes<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> grafana<span class=\"token punctuation\">&#8211;<\/span>storage<span class=\"token punctuation\">:<\/span>\/var\/lib\/grafana<br \/>\n    <span class=\"token key atrule\">depends_on<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> prometheus<\/p>\n<p><span class=\"token key atrule\">volumes<\/span><span class=\"token punctuation\">:<\/span><br \/>\n  <span class=\"token key atrule\">grafana-storage<\/span><span class=\"token punctuation\">:<\/span><\/p>\n<p>\u5728\u540c\u4e00\u76ee\u5f55\u4e0b\u521b\u5efaPrometheus\u914d\u7f6e&#xff1a;<\/p>\n<p><span class=\"token function\">mkdir<\/span> <span class=\"token parameter variable\">-p<\/span> prometheus<\/p>\n<p>\u521b\u5efaprometheus\/prometheus.yml&#xff1a;<\/p>\n<p><span class=\"token key atrule\">global<\/span><span class=\"token punctuation\">:<\/span><br \/>\n  <span class=\"token key atrule\">scrape_interval<\/span><span class=\"token punctuation\">:<\/span> 15s<br \/>\n  <span class=\"token key atrule\">evaluation_interval<\/span><span class=\"token punctuation\">:<\/span> 15s<\/p>\n<p><span class=\"token key atrule\">scrape_configs<\/span><span class=\"token punctuation\">:<\/span><br \/>\n  <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token key atrule\">job_name<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token string\">&#039;dcgm-exporter&#039;<\/span><br \/>\n    <span class=\"token key atrule\">static_configs<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token key atrule\">targets<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;dcgm-exporter:9400&#039;<\/span><span class=\"token punctuation\">]<\/span><\/p>\n<p>  <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token key atrule\">job_name<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token string\">&#039;prometheus&#039;<\/span><br \/>\n    <span class=\"token key atrule\">static_configs<\/span><span class=\"token punctuation\">:<\/span><br \/>\n      <span class=\"token punctuation\">&#8211;<\/span> <span class=\"token key atrule\">targets<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;prometheus:9090&#039;<\/span><span class=\"token punctuation\">]<\/span><\/p>\n<p>\u542f\u52a8\u6574\u4e2a\u76d1\u63a7\u6808&#xff1a;<\/p>\n<p><span class=\"token function\">docker-compose<\/span> up <span class=\"token parameter variable\">-d<\/span><\/p>\n<h3>\u6545\u969c\u6392\u9664&#xff1a;<\/h3>\n<li>DCGM Exporter\u65e0\u6cd5\u542f\u52a8&#xff1a;\n<ul>\n<li>\u68c0\u67e5NVIDIA\u9a71\u52a8\u662f\u5426\u6b63\u786e\u5b89\u88c5&#xff1a;nvidia-smi<\/li>\n<li>\u786e\u4fddDCGM\u670d\u52a1\u6b63\u5728\u8fd0\u884c&#xff1a;systemctl status nvidia-dcgm<\/li>\n<\/ul>\n<\/li>\n<li>Prometheus\u65e0\u6cd5\u6293\u53d6DCGM\u6307\u6807&#xff1a;\n<ul>\n<li>\u68c0\u67e5\u7f51\u7edc\u8fde\u63a5&#xff1a;curl localhost:9400\/metrics<\/li>\n<li>\u68c0\u67e5Prometheus\u914d\u7f6e\u4e2d\u7684\u76ee\u6807\u5730\u5740\u662f\u5426\u6b63\u786e<\/li>\n<\/ul>\n<\/li>\n<li>Grafana\u65e0\u6cd5\u8fde\u63a5Prometheus&#xff1a;\n<ul>\n<li>\u786e\u4fddPrometheus\u6b63\u5728\u8fd0\u884c&#xff1a;docker ps | grep prometheus<\/li>\n<li>\u68c0\u67e5Grafana\u4e2d\u914d\u7f6e\u7684Prometheus URL\u662f\u5426\u6b63\u786e<\/li>\n<\/ul>\n<\/li>\n<li>\u5bfc\u5165\u7684\u9762\u677f\u65e0\u6570\u636e&#xff1a;\n<ul>\n<li>\u5728Prometheus\u754c\u9762\u4e2d\u68c0\u67e5\u67e5\u8be2\u662f\u5426\u8fd4\u56de\u6570\u636e<\/li>\n<li>\u68c0\u67e5\u65f6\u95f4\u8303\u56f4\u8bbe\u7f6e\u662f\u5426\u5408\u9002<\/li>\n<\/ul>\n<\/li>\n<h3>\u6269\u5c55\u548c\u4f18\u5316&#xff1a;<\/h3>\n<li>\u6dfb\u52a0\u544a\u8b66&#xff1a;\n<ul>\n<li>\u5728Grafana\u4e2d\u4e3a\u5173\u952e\u6307\u6807\u8bbe\u7f6e\u544a\u8b66&#xff0c;\u5982\u9ad8\u6e29\u6216\u5185\u5b58\u4f7f\u7528\u8fc7\u9ad8<\/li>\n<li>\u914d\u7f6ePrometheus AlertManager\u8fdb\u884c\u66f4\u590d\u6742\u7684\u544a\u8b66\u7ba1\u7406<\/li>\n<\/ul>\n<\/li>\n<li>\u6dfb\u52a0\u66f4\u591a\u76d1\u63a7\u76ee\u6807&#xff1a;\n<ul>\n<li>\u4fee\u6539Prometheus\u914d\u7f6e\u4ee5\u76d1\u63a7\u591a\u53f0\u670d\u52a1\u5668\u7684GPU<\/li>\n<li>\u4f7f\u7528\u670d\u52a1\u53d1\u73b0\u81ea\u52a8\u6dfb\u52a0\u65b0\u7684GPU\u8282\u70b9<\/li>\n<\/ul>\n<\/li>\n<li>\u6301\u4e45\u5316\u6570\u636e&#xff1a;\n<ul>\n<li>\u914d\u7f6ePrometheus\u7684\u5b58\u50a8\u76ee\u5f55\u4ee5\u6301\u4e45\u5316\u5386\u53f2\u6570\u636e<\/li>\n<li>\u8003\u8651\u4f7f\u7528\u8fdc\u7a0b\u5b58\u50a8\u89e3\u51b3\u65b9\u6848\u7528\u4e8e\u957f\u671f\u6570\u636e\u4fdd\u7559<\/li>\n<\/ul>\n<\/li>\n<li>\u5b89\u5168\u52a0\u56fa&#xff1a;\n<ul>\n<li>\u4e3aPrometheus\u548cGrafana\u914d\u7f6e\u8eab\u4efd\u9a8c\u8bc1<\/li>\n<li>\u4f7f\u7528\u53cd\u5411\u4ee3\u7406\u548cHTTPS\u52a0\u5bc6\u8bbf\u95ee<\/li>\n<\/ul>\n<\/li>\n<p>\u901a\u8fc7\u4ee5\u4e0a\u6b65\u9aa4&#xff0c;\u60a8\u5df2\u7ecf\u6210\u529f\u90e8\u7f72\u4e86\u4e00\u4e2a\u57fa\u4e8eDCGM\u3001Prometheus\u548cGrafana\u7684GPU\u76d1\u63a7\u7cfb\u7edf&#xff0c;\u53ef\u4ee5\u5b9e\u65f6\u76d1\u63a7GPU\u7684\u5404\u9879\u5173\u952e\u6307\u6807&#xff0c;\u4e3aGPU\u8d44\u6e90\u7ba1\u7406\u548c\u4f18\u5316\u63d0\u4f9b\u6709\u529b\u652f\u6301\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u6587\u7ae0\u6d4f\u89c8\u9605\u8bfb1.4k\u6b21\uff0c\u70b9\u8d5e30\u6b21\uff0c\u6536\u85cf21\u6b21\u3002Grafana + Prometheus + DCGM\uff1aGPU\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357_prometheus\u76d1\u63a7gpu<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[1647,678,43,44],"topic":[],"class_list":["post-21293","post","type-post","status-publish","format-standard","hentry","category-server","tag-grafana","tag-prometheus","tag-43","tag-44"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v20.3 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>\u670d\u52a1\u5668GPU\u96c6\u7fa4\u786c\u4ef6\u8d44\u6e90\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.wsisp.com\/helps\/21293.html\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"\u670d\u52a1\u5668GPU\u96c6\u7fa4\u786c\u4ef6\u8d44\u6e90\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"og:description\" content=\"\u6587\u7ae0\u6d4f\u89c8\u9605\u8bfb1.4k\u6b21\uff0c\u70b9\u8d5e30\u6b21\uff0c\u6536\u85cf21\u6b21\u3002Grafana + Prometheus + DCGM\uff1aGPU\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357_prometheus\u76d1\u63a7gpu\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.wsisp.com\/helps\/21293.html\" \/>\n<meta property=\"og:site_name\" content=\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"article:published_time\" content=\"2025-04-18T20:26:48+00:00\" \/>\n<meta name=\"author\" content=\"admin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/21293.html\",\"url\":\"https:\/\/www.wsisp.com\/helps\/21293.html\",\"name\":\"\u670d\u52a1\u5668GPU\u96c6\u7fa4\u786c\u4ef6\u8d44\u6e90\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"isPartOf\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\"},\"datePublished\":\"2025-04-18T20:26:48+00:00\",\"dateModified\":\"2025-04-18T20:26:48+00:00\",\"author\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/21293.html#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.wsisp.com\/helps\/21293.html\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/21293.html#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/www.wsisp.com\/helps\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"\u670d\u52a1\u5668GPU\u96c6\u7fa4\u786c\u4ef6\u8d44\u6e90\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\",\"url\":\"https:\/\/www.wsisp.com\/helps\/\",\"name\":\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"description\":\"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\",\"name\":\"admin\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"contentUrl\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"caption\":\"admin\"},\"sameAs\":[\"http:\/\/wp.wsisp.com\"],\"url\":\"https:\/\/www.wsisp.com\/helps\/author\/admin\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"\u670d\u52a1\u5668GPU\u96c6\u7fa4\u786c\u4ef6\u8d44\u6e90\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.wsisp.com\/helps\/21293.html","og_locale":"zh_CN","og_type":"article","og_title":"\u670d\u52a1\u5668GPU\u96c6\u7fa4\u786c\u4ef6\u8d44\u6e90\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","og_description":"\u6587\u7ae0\u6d4f\u89c8\u9605\u8bfb1.4k\u6b21\uff0c\u70b9\u8d5e30\u6b21\uff0c\u6536\u85cf21\u6b21\u3002Grafana + Prometheus + DCGM\uff1aGPU\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357_prometheus\u76d1\u63a7gpu","og_url":"https:\/\/www.wsisp.com\/helps\/21293.html","og_site_name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","article_published_time":"2025-04-18T20:26:48+00:00","author":"admin","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"admin","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"3 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.wsisp.com\/helps\/21293.html","url":"https:\/\/www.wsisp.com\/helps\/21293.html","name":"\u670d\u52a1\u5668GPU\u96c6\u7fa4\u786c\u4ef6\u8d44\u6e90\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","isPartOf":{"@id":"https:\/\/www.wsisp.com\/helps\/#website"},"datePublished":"2025-04-18T20:26:48+00:00","dateModified":"2025-04-18T20:26:48+00:00","author":{"@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41"},"breadcrumb":{"@id":"https:\/\/www.wsisp.com\/helps\/21293.html#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.wsisp.com\/helps\/21293.html"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.wsisp.com\/helps\/21293.html#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.wsisp.com\/helps"},{"@type":"ListItem","position":2,"name":"\u670d\u52a1\u5668GPU\u96c6\u7fa4\u786c\u4ef6\u8d44\u6e90\u76d1\u63a7\u5b8c\u6574\u90e8\u7f72\u6307\u5357"}]},{"@type":"WebSite","@id":"https:\/\/www.wsisp.com\/helps\/#website","url":"https:\/\/www.wsisp.com\/helps\/","name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","description":"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41","name":"admin","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/","url":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","contentUrl":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","caption":"admin"},"sameAs":["http:\/\/wp.wsisp.com"],"url":"https:\/\/www.wsisp.com\/helps\/author\/admin"}]}},"_links":{"self":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/21293","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/comments?post=21293"}],"version-history":[{"count":0,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/21293\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/media?parent=21293"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/categories?post=21293"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/tags?post=21293"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/topic?post=21293"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}