This document describes the quota metrics for regional APIs and the limit
for each metric. It also describes how to update your Cloud Monitoring
dashboards, alerting policies and queries to use the regional metrics.

## Regional API quota metrics

Compute Engine regional APIs that used the global quota metrics are
migrated to use the regional metrics. Global APIs continue to use the global
metrics and quota limits. See [API rate limits for global metrics](https://docs.cloud.google.com/compute/api-quota#api-rate-limits-global).

For more information about the methods that use the regional metrics and the
limits, see [API rate limits for regional metrics](https://docs.cloud.google.com/compute/api-quota#api-rate-limits-regional).

The following table lists all metrics that are migrated from the global API quota
metrics to regional metrics:

| **Global API quota metric** | **Regional API quota metric** |
|---|---|
| `compute.googleapis.com/default` | `compute.googleapis.com/default_per_region` |
| `compute.googleapis.com/read_requests` | `compute.googleapis.com/read_requests_per_region` |
| `compute.googleapis.com/list_requests` | `compute.googleapis.com/list_requests_per_region` |
| `compute.googleapis.com/operation_read_requests` | `compute.googleapis.com/operation_read_requests_per_region` |
| `compute.googleapis.com/heavy_weight_read_requests` | `compute.googleapis.com/heavy_weight_read_requests_per_region` |
| `compute.googleapis.com/heavy_weight_write_requests` | `compute.googleapis.com/heavy_weight_write_requests_per_region` |
| `compute.googleapis.com/global_resource_write_requests` | `compute.googleapis.com/global_resource_write_requests_per_region` |

The following metrics don't have any global APIs. If you are referencing these
metrics in the Cloud Monitoring dashboards, alerts, or queries, you *must*
migrate these metrics to the corresponding regional metrics.

| **Global API quota metric** | **Regional API quota metric** |
|---|---|
| `compute.googleapis.com/simulate_maintenance_event_requests` | `compute.googleapis.com/simulate_maintenance_event_requests_per_region` |
| `compute.googleapis.com/instance_list_referrers_requests` | `compute.googleapis.com/instance_list_referrers_requests_per_region` |
| `compute.googleapis.com/get_serial_port_output_requests` | `compute.googleapis.com/get_serial_port_output_requests_per_region` |
| `compute.googleapis.com/network_endpoint_write_requests` | `compute.googleapis.com/network_endpoint_write_requests_per_region` |
| `compute.googleapis.com/network_endpoint_list_requests` | `compute.googleapis.com/network_endpoint_list_requests_per_region` |

The following metrics don't have any regional or zonal APIs. Therefore, no
changes are required for regional API quota migration.

- `compute.googleapis.com/project_set_common_instance_metadata_requests`
- `compute.googleapis.com/license_insert_requests`

## Migrate Cloud Monitoring metrics

When you migrate from the existing API quota metrics to the regional API quota
metrics, you must manually update the Cloud Monitoring dashboards, alerts, and
queries to use regional metrics.

### Update Cloud Monitoring dashboard to use the regional metrics

To update a Cloud Monitoring dashboard that monitors a global API quota
metric to a regional metric, do the following:

1. In the Google Cloud console, select **Monitoring**.
2. In the navigation pane, select **Dashboards**.
3. Select the dashboard that you want to edit to use a regional metric.
4. Click **Edit dashboard**.
5. Edit the filter that corresponds to the metric that you want to migrate:

   1. In the **Filter** field, select `quota_metric`.
   2. In the **Comparator** field, select `=~`.
   3. In the **Value** field, specify the metric name for the regional API quota.
      For example, the filter `quota_metric =~ compute.googleapis.com/read_requests.*`
      uses a regular expression to match all time-series data that uses the quota
      metric `compute.googleapis.com/read_requests.*`.

      ![Edit filter value to use a regional metric.](https://docs.cloud.google.com/static/compute/images/dashboard-metric.png)

      For more information about the metrics, see [API rate limits for regional metrics](https://docs.cloud.google.com/compute/api-quota#api-rate-limits-regional).
   4. Click **Done** to save changes.

   5. Optional: If you want to view the usage of quota per region,
      expand the **Grouped** header and select `location` in the
      **Group by** field.

6. Click **Close editor**.

### Update Cloud Monitoring alerting policies to use the regional API quota metrics

To change an alerting policy that monitors a global API quota metric to a
regional metric, do the following:

1. In the Google Cloud console, select **Monitoring**.
2. In the navigation pane, select **Alerting**. The Alerting page displays panes that list summary information, incidents, and alerting policies.
3. Click the alerting policy that you want to edit to use a regional metric.
4. Click **Edit** to modify the policy.
5. Edit the filter that corresponds to the metric that you want to migrate:

   1. In the **Filter** field, select `quota_metric`.
   2. In the **Comparator** field, select `=~`.
   3. In the **Value** field, specify the metric name for the regional API quota.
      For example, the filter `quota_metric =~ compute.googleapis.com/read_requests.*`
      uses a regular expression to match all time-series data that uses the quota
      metric `compute.googleapis.com/read_requests.*`.

      ![Edit alert to use a regional metric.](https://docs.cloud.google.com/static/compute/docs/images/alert-metric.png)

      For more information about the metrics, see [API rate limits for regional metrics](https://docs.cloud.google.com/compute/api-quota#api-rate-limits-regional).
   4. Click **Done** to save changes in the filter.

   5. Optional: If you want a per-region alert for the regional metric,
      expand the **Across time series** header and select `location` in the
      **Time series group by** field.

6. Click **Save policy**.

### Update Cloud Monitoring queries to use the regional quota metrics

To change the quota metric in a PromQL-based alerting policy or a chart in
a dashboard, use the code editor and edit the query filter. See
[Using the code editor for PromQL](https://docs.cloud.google.com/monitoring/promql/promql-in-monitoring).

Open the code editor and edit the query as shown in the following example:

    rate({
       "__name__"="serviceruntime.googleapis.com/quota/rate/net_usage",
       "monitored_resource"="consumer_quota",
       "quota_metric"=~"compute.googleapis.com/read_requests.*"
    }[${__interval}])

Optionally, you can group the results based on a location by using the
`sum by` operation:

    sum by (location) (
       rate({
           "__name__"="serviceruntime.googleapis.com/quota/rate/net_usage",
           "monitored_resource"="consumer_quota",
           "quota_metric"=~"compute.googleapis.com/read_requests.*"
       }[${__interval}])
    )