Cloud Monitoring lets you evaluate system performance and health across
your infrastructure using data aggregation, custom metrics, filtering, and
metrics scopes. Understanding these foundational concepts helps you
structure cross-project observability and isolate specific signals from your
time-series data.

## Aggregation

When you read metric data using the Metrics Explorer or the Monitoring API,
you can use *aggregation* to summarize the time series data. Aggregation
typically starts with an alignment step in which each time series' data is
placed on the same time boundaries. Next, a new time series is created by
combining the data points from multiple time series, using operations like
average, sum, minimum, maximum, and so forth.

For more information about aggregation, see the
[`projects.timeSeries.list`](https://docs.cloud.google.com/monitoring/api/ref_v3/rest/v3/projects.timeSeries/list)
API method.

## Custom metrics

Cloud Monitoring has a large number of built-in system metrics, but you can
also create custom metrics. You must create a metric descriptor that
describes your custom metric and one or more time series that contain its data.
You can use your custom metric data in charts and alerts. You can also create
custom metrics based on logs data. For more information, see the following
pages:

- [Metrics, Time Series, and Resources](https://docs.cloud.google.com/monitoring/api/v3/metrics) provides an overview.
- [Using Custom Metrics](https://docs.cloud.google.com/monitoring/custom-metrics) shows you how to create, write, and read custom metrics.
- [Logs-based Metrics](https://docs.cloud.google.com/logging/docs/view/logs_based_metrics) describes how to create custom metrics from logs.
- [Monitoring API](https://docs.cloud.google.com/monitoring/api/ref_v3/rest), is the API used to for custom metrics. See the `projects.metricDescriptors` and `projects.timeSeries` collections.

## Filtering

Cloud Logging and Cloud Monitoring provide different kinds of filters
that let you select
sets of items. Filters are strings that contain combinations of comparisons
appropriate to their kind. For example:

- [Logging filters](https://docs.cloud.google.com/logging/docs/view/advanced_filters) let you select particular log entries based on log names, where the logs came from, their payload content, and so on.
- [Monitoring filters](https://docs.cloud.google.com/monitoring/api/v3/filters) let you select particular metric descriptors, resource groups, and time series data.
- [Trace filters](https://docs.cloud.google.com/trace/docs/trace-filters) let you select traces based on span names, latency, and label values.

## Metrics scope

Monitoring organizes itself around metrics scopes that
let you view and monitor metrics for one or more Google Cloud projects.
For more information about metrics scopes, see
[Configuring your Google Cloud project for Cloud Monitoring](https://docs.cloud.google.com/monitoring/settings).