This is an automated email from the ASF dual-hosted git repository.
gongchao pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/hertzbeat.git
The following commit(s) were added to refs/heads/master by this push:
new cb11f11ac [feature] add NVIDIA monitor (#2643)
cb11f11ac is described below
commit cb11f11ace5ceae9fce3ed67a4193c24d3771833
Author: Jast <[email protected]>
AuthorDate: Tue Sep 3 23:38:53 2024 +0800
[feature] add NVIDIA monitor (#2643)
Co-authored-by: aias00 <[email protected]>
Co-authored-by: shown <[email protected]>
Co-authored-by: Calvin <[email protected]>
Co-authored-by: tomsun28 <[email protected]>
---
home/docs/help/nvidia.md | 37 ++++
.../current/help/nvidia.md | 37 ++++
manager/src/main/resources/define/app-nvidia.yml | 198 +++++++++++++++++++++
3 files changed, 272 insertions(+)
diff --git a/home/docs/help/nvidia.md b/home/docs/help/nvidia.md
new file mode 100644
index 000000000..6cc3f0d2f
--- /dev/null
+++ b/home/docs/help/nvidia.md
@@ -0,0 +1,37 @@
+---
+id: nvidia
+title: NVIDIA Monitoring
+sidebar_label: NVIDIA Monitoring
+keywords: [Open Source Monitoring System, NVIDIA Monitoring]
+---
+
+> Collect and monitor general performance metrics of NVIDIA operating systems.
+> NVIDIA monitoring requires the nvidia-smi command, which is installed
together with the NVIDIA GPU driver. So when monitoring NVIDIA, we need to
install the NVIDIA GPU driver.
+
+### Configuration Parameters
+
+| Parameter Name | Description
|
+|------------------|-------------------------------------------------------------|
+| Monitoring Host | The IP address (IPv4/IPv6) or domain name of the
monitored endpoint. Note ⚠️ do not include protocol headers (e.g., https://,
http://). |
+| Task Name | The name identifying this monitoring task, which needs to
be unique. |
+| Port | The port exposed for Linux SSH, default is 22.
|
+| Username | SSH connection username, optional.
|
+| Password | SSH connection password, optional.
|
+| Collection Interval | Interval for periodically collecting monitoring data,
in seconds. The minimum interval is 30 seconds. |
+| Probe Before Monitoring | Whether to probe the monitoring endpoint to check
its availability before adding it. Monitoring is added or modified only if the
probe succeeds. |
+| Description/Remarks | Additional notes and descriptions for this monitoring
task. Users can add relevant information here. |
+
+### Collected Metrics
+
+#### Metric Set: basic
+
+| Metric Name | Unit | Description |
+|------------------------|--------|------------------|
+| index | None | GPU index |
+| name | None | GPU name |
+| utilization.gpu[%] | None | GPU utilization |
+| utilization.memory[%] | None | Memory utilization |
+| memory.total[MiB] | MiB | Total memory |
+| memory.used[MiB] | MiB | Used memory |
+| memory.free[MiB] | MiB | Free memory |
+| temperature.gpu | None | GPU temperature |
diff --git
a/home/i18n/zh-cn/docusaurus-plugin-content-docs/current/help/nvidia.md
b/home/i18n/zh-cn/docusaurus-plugin-content-docs/current/help/nvidia.md
new file mode 100644
index 000000000..8e3f190ad
--- /dev/null
+++ b/home/i18n/zh-cn/docusaurus-plugin-content-docs/current/help/nvidia.md
@@ -0,0 +1,37 @@
+---
+id: nvidia
+title: 监控：NVIDIA 监控
+sidebar_label: NVIDIA 监控
+keywords: [开源监控系统, NVIDIA监控]
+---
+
+> 对 NVIDIA 操作系统的通用性能指标进行采集监控。
+> NVIDIA 监控需要用到 nvidia-smi 命令，nvidia-smi 是与 NVIDIA GPU 驱动程序一起安装的。所以在监控 NVIDIA
时，我们需要安装 NVIDIA GPU 驱动程序。
+
+### 配置参数
+
+| 参数名称 | 参数帮助描述 |
+|--------|---------------------------------------------------------|
+| 监控Host | 被监控的对端 IPV4，IPV6 或 域名。注意⚠️不带协议头(eg: https://, http://)。 |
+| 任务名称 | 标识此监控的名称，名称需要保证唯一性。 |
+| 端口 | Linux SSH 对外提供的端口，默认为22。 |
+| 用户名 | SSH 连接用户名，可选 |
+| 密码 | SSH 连接密码，可选 |
+| 采集间隔 | 监控周期性采集数据间隔时间，单位秒，可设置的最小间隔为30秒 |
+| 是否探测 | 新增监控前是否先探测检查监控可用性，探测成功才会继续新增修改操作 |
+| 描述备注 | 更多标识和描述此监控的备注信息，用户可以在这里备注信息 |
+
+### 采集指标
+
+#### 指标集合：basic
+
+| 指标名称 | 指标单位 | 指标帮助描述 |
+|--------------------|------|--------|
+| index | 无 | 显卡索引 |
+| name | 无 | 显卡名称 |
+| utilization.gpu[%] | 无 | GPU利用率 |
+| utilization.memory[%] | 无 | 显存利用率 |
+| memory.total[MiB] | 无 | 总显存 |
+| memory.used[MiB] | 无 | 已用显存 |
+| memory.free[MiB] | 无 | 空闲显存 |
+| temperature.gpu | 无 | 显卡温度 |
diff --git a/manager/src/main/resources/define/app-nvidia.yml
b/manager/src/main/resources/define/app-nvidia.yml
new file mode 100644
index 000000000..583f2189b
--- /dev/null
+++ b/manager/src/main/resources/define/app-nvidia.yml
@@ -0,0 +1,198 @@
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements. See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# The monitoring type category：service-application service monitoring
db-database monitoring custom-custom monitoring os-operating system monitoring
+category: server
+# The monitoring type eg: linux windows tomcat mysql aws...
+app: nvidia
+# The monitoring i18n name
+name:
+ zh-CN: NVIDIA
+ en-US: NVIDIA
+# The description and help of this monitoring type
+help:
+ zh-CN: Hertzbeat 使用 <a class='help_module_content'
href='https://hertzbeat.apache.org/docs/advanced/extend-ssh'> SSH 协议</a> 对
NVIDIA GPU显卡的通用性能指标进行采集监控。<br>您可以点击“<i>新建
NVIDIA</i>”并配置HOST端口账户等相关参数进行添加，支持SSH账户密码或密钥认证。或者选择“<i>更多操作</i>”，导入已有配置。
+ en-US: Hertzbeat uses the <a class='help_module_content'
href='https://hertzbeat.apache.org/docs/advanced/extend-ssh'>SSH protocol</a>
to collect and monitor general performance metrics of NVIDIA GPUs. <br> You can
click " <i>Create NVIDIA</i> " to add and configure parameters such as HOST,
port, account, etc., supporting SSH account password or key authentication.
Alternatively, you can select " <i>More Actions</i> " to import an existing
configuration.
+ zh-TW: Hertzbeat 使用 <a class='help_module_content'
href='https://hertzbeat.apache.org/docs/advanced/extend-ssh'>SSH 協議</a> 對
NVIDIA GPU 顯卡的通用性能指標進行採集監控。<br>您可以點擊“<i>新建 NVIDIA</i>”並配置
HOST、端口、帳戶等相關參數進行添加，支持 SSH 帳戶密碼或密鑰認證。或者選擇“<i>更多操作</i>”，導入已有配置。
+helpLink:
+ zh-CN: https://hertzbeat.apache.org/zh-cn/docs/help/nvidia/
+ en-US: https://hertzbeat.apache.org/docs/help/nvidia/
+# Input params define for monitoring(render web ui by the definition)
+params:
+ # field-param field key
+ - field: host
+ # name-param field display i18n name
+ name:
+ zh-CN: 目标Host
+ en-US: Target Host
+ # type-param field type(most mapping the html input type)
+ type: host
+ # required-true or false
+ required: true
+ # field-param field key
+ - field: port
+ # name-param field display i18n name
+ name:
+ zh-CN: 端口
+ en-US: Port
+ # type-param field type(most mapping the html input type)
+ type: number
+ # when type is number, range is required
+ range: '[0,65535]'
+ # required-true or false
+ required: true
+ # default value
+ defaultValue: 22
+ # field-param field key
+ - field: timeout
+ # name-param field display i18n name
+ name:
+ zh-CN: 超时时间(ms)
+ en-US: Timeout(ms)
+ # type-param field type(most mapping the html input type)
+ type: number
+ # when type is number, range is required
+ range: '[400,200000]'
+ # required-true or false
+ required: false
+ # default value
+ # 默认值
+ defaultValue: 6000
+ # field-param field key
+ - field: reuseConnection
+ # name-param field display i18n name
+ name:
+ zh-CN: 复用连接
+ en-US: Reuse Connection
+ # type-param field type(most mapping the html input type)
+ type: boolean
+ # required-true or false
+ required: true
+ defaultValue: false
+ # field-param field key
+ - field: username
+ # name-param field display i18n name
+ name:
+ zh-CN: 用户名
+ en-US: Username
+ # type-param field type(most mapping the html input type)
+ type: text
+ # when type is text, use limit to limit string length
+ limit: 50
+ # required-true or false
+ required: true
+ # field-param field key
+ - field: password
+ # name-param field display i18n name
+ name:
+ zh-CN: 密码
+ en-US: Password
+ # type-param field type(most mapping the html input tag)
+ type: password
+ # required-true or false
+ required: false
+ # field-param field key
+ - field: privateKey
+ # name-param field display i18n name
+ name:
+ zh-CN: 私钥
+ en-US: PrivateKey
+ # type-param field type(most mapping the html input type)
+ type: textarea
+ placeholder: -----BEGIN RSA PRIVATE KEY-----
+ # required-true or false
+ required: false
+ # hide param-true or false
+ hide: true
+# collect metrics config list
+metrics:
+ # metrics - basic, inner monitoring metrics (responseTime - response time)
+ - name: basic
+ i18n:
+ zh-CN: 显卡基本信息
+ en-US: Basic Information
+ # metrics scheduling priority(0->127)->(high->low), metrics with the same
priority will be scheduled in parallel
+ # priority 0's metrics is availability metrics, it will be scheduled
first, only availability metrics collect success will the scheduling continue
+ priority: 0
+ # collect metrics content
+ fields:
+ # field-metric name, type-metric type(0-number,1-string), unit-metric
unit('%','ms','MB'), label-whether it is a metrics label field
+ - field: index
+ type: 1
+ label: true
+ i18n:
+ zh-CN: 显卡索引
+ en-US: Host Name
+ - field: name
+ type: 1
+ i18n:
+ zh-CN: 显卡名称
+ en-US: System Version
+ - field: utilization.gpu [%]
+ type: 0
+ unit: '%'
+ i18n:
+ zh-CN: GPU利用率
+ en-US: GPU Utilization
+ - field: utilization.memory [%]
+ type: 0
+ unit: '%'
+ i18n:
+ zh-CN: 显存利用率
+ en-US: Memory Utilization
+ - field: memory.total [MiB]
+ type: 1
+ unit: 'MiB'
+ i18n:
+ zh-CN: 总显存
+ en-US: Total Memory
+ - field: memory.used [MiB]
+ type: 0
+ unit: 'MiB'
+ i18n:
+ zh-CN: 已用显存
+ en-US: Used Memory
+ - field: memory.free [MiB]
+ type: 0
+ unit: 'MiB'
+ i18n:
+ zh-CN: 空闲显存
+ en-US: Free Memory
+ - field: temperature.gpu
+ type: 1
+ unit: '°C'
+ i18n:
+ zh-CN: 显卡温度
+ en-US: GPU Temperature
+ # the protocol used for monitoring, eg: sql, ssh, http, telnet, wmi, snmp,
sdk
+ protocol: ssh
+ # the config content when protocol is ssh
+ ssh:
+ # ssh host: ipv4 ipv6 domain
+ host: ^_^host^_^
+ # ssh port
+ port: ^_^port^_^
+ # ssh username
+ username: ^_^username^_^
+ # ssh password
+ password: ^_^password^_^
+ # ssh private key
+ privateKey: ^_^privateKey^_^
+ timeout: ^_^timeout^_^
+ reuseConnection: ^_^reuseConnection^_^
+ # ssh run collect script
+ script: nvidia-smi
--query-gpu=index,name,utilization.gpu,utilization.memory,memory.total,memory.used,memory.free,temperature.gpu
--format=csv,nounits | sed 's/ *, */,/g' | sed 's/ / /g' | sed 's/,/ /g'
+ # ssh response data parse type: oneRow, multiRow
+ parseType: multiRow
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
