This topic describes how to view the monitoring data of graphics processing units (GPUs) in the CloudMonitor console and query the data by calling an API operation.

Prerequisites

GPU metrics

GPU metrics can be viewed based on GPUs, instances, and application groups. The following table describes the GPU metrics.
Metric Unit MetricName Dimensions
(Agent) gpu decoder utilization % gpu_decoder_utilization userId, instanceId, and gpuId
(Agent) gpu encoder utilization % gpu_encoder_utilization userId, instanceId, and gpuId
(Agent) gpu gpu temperature °C gpu_gpu_temperature userId, instanceId, and gpuId
(Agent) gpu gpu usedutilization % gpu_memory_userdutilization userId, instanceId, and gpuId
(Agent) gpu memory freespace Byte gpu_memory_freespace userId, instanceId, and gpuId
(Agent) gpu memory freeutilization % gpu_memory_freeutilization userId, instanceId, and gpuId
(Agent) gpu memory usedspace Byte gpu_memory_userdspace userId, instanceId, and gpuId
(Agent) gpu memory usedutilization % gpu_memory_usedutilization userId, instanceId, and gpuId
(Agent) gpu power readings power draw W gpu_power_readings_power_draw userId, instanceId, and gpuId

View the monitoring data of the GPU by using the CloudMonitor console

  1. Log on to the CloudMonitor console.
  2. In the left-side navigation pane, click Host Monitoring.
  3. On the Instances tab of the Host Monitoring page, click the instance name of the required host or click Monitoring Charts in the Actions column of the required host.
  4. On the Instance Info page, click the GPUMonitor tab.
    On the GPUMonitor tab, view the monitoring charts of the GPU.

Query the monitoring data of the GPU by calling an API operation

You can call the DescribeMetricList to query the monitoring data of a cloud service. For more information, see DescribeMetricList.
Note Set the Namespace parameter to acs_ecs_dashboard and the MetricName parameter to Dimensions. For more information, see GPU metrics.