ProjectedMetric - AWS Compute Optimizer

ProjectedMetric

Describes a projected utilization metric of a recommendation option, such as an Amazon EC2 instance. This represents the projected utilization of a recommendation option had you used that resource during the analyzed period.

Compare the utilization metric data of your resource against its projected utilization metric data to determine the performance difference between your current resource and the recommended option.

Note

The Cpu, Memory, GPU, and GPU_MEMORY metrics are the only projected utilization metrics returned when you run the GetEC2RecommendationProjectedMetrics action. Additionally, these metrics are only returned for resources with the unified CloudWatch agent installed on them. For more information, see Enabling Memory Utilization with the CloudWatch Agent and Enabling NVIDIA GPU utilization with the CloudWatch Agent.

Contents

name

The name of the projected utilization metric.

The following projected utilization metrics are returned:

  • Cpu - The projected percentage of allocated EC2 compute units that would be in use on the recommendation option had you used that resource during the analyzed period. This metric identifies the processing power required to run an application on the recommendation option.

    Depending on the instance type, tools in your operating system can show a lower percentage than CloudWatch when the instance is not allocated a full processor core.

  • Memory - The percentage of memory that would be in use on the recommendation option had you used that resource during the analyzed period. This metric identifies the amount of memory required to run an application on the recommendation option.

    Units: Percent

    Note

    The Memory metric is only returned for resources with the unified CloudWatch agent installed on them. For more information, see Enabling Memory Utilization with the CloudWatch Agent.

  • GPU - The projected percentage of allocated GPUs if you adjust your configurations to Compute Optimizer's recommendation option.

  • GPU_MEMORY - The projected percentage of total GPU memory if you adjust your configurations to Compute Optimizer's recommendation option.

    Note

    The GPU and GPU_MEMORY metrics are only returned for resources with the unified CloudWatch Agent installed on them. For more information, see Enabling NVIDIA GPU utilization with the CloudWatch Agent.

Type: String

Valid Values: Cpu | Memory | EBS_READ_OPS_PER_SECOND | EBS_WRITE_OPS_PER_SECOND | EBS_READ_BYTES_PER_SECOND | EBS_WRITE_BYTES_PER_SECOND | DISK_READ_OPS_PER_SECOND | DISK_WRITE_OPS_PER_SECOND | DISK_READ_BYTES_PER_SECOND | DISK_WRITE_BYTES_PER_SECOND | NETWORK_IN_BYTES_PER_SECOND | NETWORK_OUT_BYTES_PER_SECOND | NETWORK_PACKETS_IN_PER_SECOND | NETWORK_PACKETS_OUT_PER_SECOND | GPU_PERCENTAGE | GPU_MEMORY_PERCENTAGE

Required: No

timestamps

The timestamps of the projected utilization metric.

Type: Array of timestamps

Required: No

values

The values of the projected utilization metrics.

Type: Array of doubles

Required: No

See Also

For more information about using this API in one of the language-specific AWS SDKs, see the following: