腾讯云可观测平台 资源管理(腾讯云大模型训推平台 TI-ONE)

这里说明了腾讯云资源管理指标监控数据的查询方法,包括命名空间、监控指标及各维度对应参数的说明。查询资源管理指标监控数据: 用户可以通过设置特定的命名空间和维度参数来查询相关的监控数据。命名空间为QCE/TI_RESOURCE,维度参数包括AppId(账号ID)、ResourceGroupId(资源组ID)和ResourceId(资源ID)。

相关产品
腾讯云可观测平台 产品页,含版本规格、价格与购买入口。

一、命名空间

Namespace = QCE/TI_RESOURCE

二、监控指标

指标英文名 指标中文名 指标说明 单位 维度 统计粒度
GroupCpuUsage CPU 使用率 资源组 CPU 使用率 % ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupGpuUtil GPU 使用率 资源组 GPU 使用率 % ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupLanInTraffic 内网入带宽 资源组内网入带宽 Mbps ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupLanOutTraffic 内网出宽带 资源组内网出宽带 Mbps ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupMemUsage 内存使用率 资源组内存使用率 % ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupWanInTraffic 外网入带宽 资源组外网入带宽 Mbps ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupWanOutTraffic 外网出宽带 资源组外网出宽带 Mbps ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsCpuUsage CPU 利用率 资源(节点)CPU 利用率 % ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsGpuUtil GPU 使用率 资源(节点)GPU 使用率 % ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsMemUsage 内存利用率 资源(节点)内存利用率 % ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsLanInTraffic 内网入带宽 资源(节点)内网入带宽 Mbps ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsLanOutTraffic 内网出带宽 资源(节点)内网出带宽 Mbps ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsWanInTraffic 外网入带宽 资源(节点)外网入带宽 Mbps ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsWanOutTraffic 外网出带宽 资源(节点)外网出带宽 Mbps ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupWanOutratio 外网带宽利用率 外网带宽利用率 % ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupFp16EngineActivity FP16活跃时间比 FP16活跃时间比 % ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupFp32EngineActivity FP32活跃时间比 FP32活跃时间比 % ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupFp64EngineActivity FP64活跃时间比 FP64活跃时间比 % ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupGpuDecUtil GPU 解码使用率 GPU 解码使用率 % ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupGpuEccPersistent 持久性 ECC 错误 持久性 ECC 错误 Count ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupGpuEccVolatile 易失性 ECC 错误 易失性 ECC 错误 Count ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupGpuEncUtil GPU 编码使用率 GPU 编码使用率 % ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupGpuPowerUsage GPU 功耗 GPU 功耗 W ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupGpuRetiredPages 停用的内存页面 停用的内存页面 Count ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupGpuTemperature GPU 温度 GPU 温度 °C ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
GroupGpuXidErrors GPUxid 计数 GPUxid 计数 Count ResourceGroupIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsWanOutratio 外网带宽利用率 外网带宽利用率 % ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsGpuDecUtil GPU 解码使用率 GPU 解码使用率 % ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsGpuEccPersistent 持久性 ECC 错误 持久性 ECC 错误 Count ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsGpuEccVolatile 易失性 ECC 错误 易失性 ECC 错误 Count ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsGpuEncUtil GPU 编码使用率 GPU 编码使用率 % ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsGpuPowerUsage GPU 功耗 GPU 功耗 W ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsGpuRetiredPages 停用的内存页面 停用的内存页面 Count ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsGpuTemperature GPU 温度 GPU 温度 °C ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsGpuXidErrors GPUxid 计数 GPUxid 计数 Count ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsFp16EngineActivity FP16活跃时间比 FP16活跃时间比 % ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsFp32EngineActivity FP32活跃时间比 FP32活跃时间比 % ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]
InsFp64EngineActivity FP64活跃时间比 FP64活跃时间比 % ResourceIdAppId [ 10s, avg ] [ 60s, avg ] [ 300s, avg ] [ 3600s, avg ] [ 86400s, avg ]

三、各维度对应参数总览

参数名称 维度名称 维度解释 格式
Instances.N.Dimensions.0.Name AppId 账号基本信息 APPID 的维度名称 输入 String 类型维度名称:AppId(SDK 调用时会自动获取,无需传参)
Instances.N.Dimensions.0.Value AppId 账号基本信息 APPID 输入 ID,例如:1231231231(SDK 调用时会自动获取,无需传参)
Instances.N.Dimensions.1.Name ResourceGroupId 资源组 ID 的维度名称 输入 String 类型维度名称:ResourceGroupId
Instances.N.Dimensions.1.Value ResourceGroupId 资源组 ID 输入 ID,例如:trsg-b564kzx2
Instances.N.Dimensions.2.Name ResourceId 资源组中的某个资源 ID 的维度名称 输入 String 类型维度名称:ResourceId
Instances.N.Dimensions.2.Value ResourceId 资源组中的某个资源 ID 输入 ID,例如:sm-54wplvsv

四、入参说明

查询资源管理指标监控数据,取值如下:

&Namespace=QCE/TI_RESOURCE

&Instances.N.Dimensions.0.Name=AppId

&Instances.N.Dimensions.0.Value=具体的账号 ID

&Instances.N.Dimensions.1.Name=ResourceGroupId

&Instances.N.Dimensions.1.Value=具体的资源组 ID

&Instances.N.Dimensions.2.Name=ResourceId

&Instances.N.Dimensions.2.Value=资源组中的某个资源 ID

赞 (0)
上一篇 4天前
下一篇 4天前