VM instance sizing recommender

The VM instance sizing recommender helps you optimize the resource utilization of your Compute Engine virtual machine instances. These recommendations are generated automatically based on system metrics or metrics gathered by the Cloud Monitoring service over the previous 8 days. You can use these recommendations to resize your instance machine types to more efficiently use VM resources.

Sizing recommendations are also known as rightsizing recommendations.

See Applying sizing recommendations for VM instances for more information.

Warning:

Before applying recommendations in the Cloud Console or using the API, ensure that impacts are assessed by a reviewer. The reviewer should have the knowledge to assess impacts identified in recommendations, as well as impacts specific to your infrastructure and business. Applying recommendations without proper assessment could result in unexpected changes, such as issues with system performance, poor reliability, or loss of required permissions. If you choose to apply recommendations without human review, ensure that you have set up a rollback process before making any changes.

Recommender ID

The VM instance sizing recommender ID is:

google.compute.instance.MachineTypeRecommender

You use this ID when you view and modify recommendations using gcloud commands, or the REST and RPC APIs.

Location

The Recommender gcloud commands and APIs require you to specify a location for recommendations that you view or modify. For VM instance sizing recommendations, use the Compute Engine zone as the location in gcloud and API interactions.

Permissions

Required permissions

To view VM instance sizing recommendations:

• recommender.computeInstanceMachineTypeRecommendations.get
• recommender.computeInstanceMachineTypeRecommendations.list

To modify state of VM instance sizing recommendations:

• recommender.computeInstanceMachineTypeRecommendations.update

Note: If you use the x-goog-user-project header or userProject query string parameter in your request, you must have serviceusage.services.use permission for the project ID that you specify, in addition to the normal Identity and Access Management (IAM) permissions required to make the request. The Owner (roles/owner) or Editor (roles/editor) role includes this permission.

Note: Granting your users these permissions might give them a partial view of the resource for which the recommendation is generated. For more information, see the Recommender Overview.

Granting Permissions

To grant these permissions, assign roles as follows:

• To view recommendations only, grant the Compute Recommender Viewer (roles/recommender.computeViewer) role.
• To view and modify recommendations, grant the Compute Recommender Admin (roles/recommender.computeAdmin) role.
• To grant the serviceusage.services.use permission, grant the Service Usage Consumer role (roles/serviceusage.serviceUsageConsumer).

As an alternative, you can also grant the following basic roles:

• To view recommendations only, grant the Viewer (roles/viewer) role.
• To view and modify recommendations, grant the Owner (roles/owner) or Editor (roles/editor) role.
  • These roles also include the serviceusage.services.use permission.

Note: To change the IAM policy for a project, you must be granted a role that includes the resourcemanager.projects.setIamPolicy permission. For more details about IAM permissions, see IAM overview.

Recommendation Subtypes

This Recommender supports the following subtypes:

• CHANGE_MACHINE_TYPE - Recommends changing the current machine type of a virtual machine instances based on current usage.

Operation groups

Each recommendation generated by the VM instance sizing recommender has an operation group that describes a set of actions that must be taken in order to apply the recommendation.

Sizing recommendation for a standalone VM instance

A sizing recommendation for a standalone VM instance includes one OperationGroup with two operations on the compute.googleapis.com/Instance resource:

1. A test operation to make sure that the current machine type is the same as when the recommendation was made. For example:

   {
    "action": "test",
    "resourceType": "compute.googleapis.com/Instance",
    "resource" : "//compute.googleapis.com/projects/example-project/zones/z1/instances/vm-name",
    "path": "/machineType",
    "valueMatcher": {
      "matchesPattern": ".*zones/us-central1-f/machineTypes/n1-standard-32"
    }
   }
   

2. A replace operation that replaces the machine type with the recommended type. For example:

    {
      "action": "replace",
      "resourceType": "compute.googleapis.com/Instance",
      "resource" : "//compute.googleapis.com/projects/example-project/zones/z1/instances/vm-name",
      "path": "/machineType",
      "value": "zones/us-central1-f/machineTypes/n1-standard-4"
    }
   

Examples

The following example shows how to list VM instance sizing recommendations:

1. Set required environment variables:

   PROJECT=TARGET_PROJECT_ID
   LOCATION=LOCATION_ID
   RECOMMENDER=google.compute.instance.MachineTypeRecommender
   

   where:

   • TARGET_PROJECT_ID is the project whose recommendations you want to list. This can be a different project than your current project.

     • For gcloud commands, you must use the project ID
     • For API requests, you can use the project number or project ID. Project number is recommended.

     The project number is returned in responses from both the API and gcloud commands.

   • LOCATION_ID is the Google Cloud location where resources associated with the recommendations are located (for example, us-central1-a).

2. List VM instance sizing recommendations:

   gcloud

   Enter the following:

   gcloud recommender recommendations list \
       --project=${PROJECT} \
       --location=${LOCATION} \
       --recommender=${RECOMMENDER} \
       --format=json
   

   REST

   Enter the following:

   curl \
       -H "Authorization: Bearer $(gcloud auth print-access-token)"  \
       -H "x-goog-user-project: ${PROJECT}" \
       "https://recommender.googleapis.com/v1/projects/${PROJECT}/locations/${LOCATION}/recommenders/${RECOMMENDER}/recommendations"
   

The output is similar to the following:

[
  {
    "content": {
     "operationGroups": [
        {
          "operations": [
            {
              "action": "test",
              "path": "/machineType",
              "resource": "//compute.googleapis.com/projects/example-project/zones/us-central1-a/instances/instance-1",
              "resourceType": "compute.googleapis.com/Instance",
              "valueMatcher": {
                "matchesPattern": ".*zones/us-central1-a/machineTypes/n1-standard-4"
              }
            },
            {
              "action": "replace",
              "path": "/machineType",
              "resource": "//compute.googleapis.com/projects/example-project/zones/us-central1-a/instances/instance-1",
              "resourceType": "compute.googleapis.com/Instance",
              "value": "zones/us-central1-a/machineTypes/custom-2-5120"
            }
          ]
        }
      ]
    },
    "description": "Save cost by changing machine type from n1-standard-4 to custom-2-5120.",
    "etag": "280b34810bba8a1a",
    "lastRefreshTime": "2019-06-28T06:49:21Z",
    "name": "projects/548293842938/locations/us-central1-a/recommenders/google.compute.instance.MachineTypeRecommender/recommendations/a523ff7e-ed03-4143-a3a5-5b396b99cba9",
    "primaryImpact": { ... }
    },
    "stateInfo": {
      "state": "ACTIVE"
    },
    "recommenderSubtype": "CHANGE_MACHINE_TYPE"
  }
]

See Using the API for instructions on performing additional tasks on recommendations using the Recommender gcloud commands and APIs.