Metrics provide visibility into the health and performance of workloads running on Akamai Cloud. Exporting those metrics to external observability platforms often requires custom integrations, ongoing maintenance, and additional operational overhead.

The Akamai Distribution of the OpenTelemetry Collector for Akamai Cloud Pulse (the collector), simplifies metric export by providing a standardized method for collecting and exporting metrics. Available under Apache License, Version 2.0, the collector retrieves metrics from supported Akamai Cloud services through the Linode API and exports them to one or more observability platforms that support the OpenTelemetry standard.

The collector is available as a compiled Linux binary, a Docker image, or a Helm chart for Kubernetes environments.

The OpenTelemetry pipeline consists of receivers, processors, and exporters. Receivers collect metrics from supported Akamai Cloud services. Processors filter, transform, or batch metrics before export. Exporters send processed metrics to one or more observability platforms where the data can be stored, queried, and visualized using monitoring and analytics tools. To learn which receivers, processors, and exporters are supported, see Supported components.

Prerequisites

A personal access token (PAT) is required to authenticate the akamaicloudpulsereceiver and allow it to access metric data through the Linode API. The PAT should be created with:

• Read access for Monitor
• Read access for the service for which metrics are being collected

Supported Akamai Cloud services

The collector is currently available for the following services:

• Managed Databases (service_type: "dbaas")
• Nodebalancers (service_type: "nodebalancer")
• Object Storage (service_type: "objectstorage")
• LKE Enterprise (service_type: "lke")
• Logs (service_type: "logs")

Support for additional services is planned for future releases.

Supported operating systems

The collector provides build artifacts specifically compiled for AMD64 Linux platforms and should be deployed only to compatible Linux environments.

The collector has been verified on the following supported Linux operating systems:

• Ubuntu 22.08+
• Debian 11+,
• AlmaLinux 8+
• Alpine 3.19+
• CentOS Stream 9
• Fedora 41
• Rocky Linux 8.

Supported components

The collector includes receivers, processors, exporters, and extensions that can be configured to support different metrics collection and export workflows.

Receivers

Receivers collect metric data from configured sources.

The following receivers are supported:

• akamaicloudpulsereceiver
• prometheusreceiver
• otlpreceiver

For information about configuring the Akamai Cloud Pulse receiver (akamaicloudpulsereceiver), including mandatory and supported settings, see Receiver configuration.

Processors

Processors modify, filter, or batch metric data before export.

The following processors are supported:

• batchprocessor
• memorylimiterprocessor
• filterprocessor
• attributesprocessor
• transformprocessor

Exporters

Exporters send metric data to external observability platforms or endpoints.

The following exporters are supported:

• otlphttpexporter
• prometheusremotewriteexporter
• prometheusexporter
• debugexporter

Extensions

Extensions provide supporting functionality for the collector.

The following extensions are supported:

• healthcheckextension
• memorylimiterextension

Sample configurations

Sample configuration files are available for common metrics export scenarios. You can find these sample configuration files in the GitHub repository.

API rate limits

Metrics collection is subject to system and API rate limits and service-specific constraints that may affect collection behavior and API request throughput. For more information, see API rate limits and request batching.

Next steps

• To get started with configuring and deploying the collector, see Quick start: metrics export.
• To learn more about receiver configuration, see Receiver configuration.