Cloudera Docs

What is Cloudera Observability and how is it useful

Cloudera Observability is a Cloudera service that helps you interactively understand your environment, data services, workloads, clusters, and resources. Its wide range of metrics and health tests help you identify and troubleshoot existing and potential problems. This service also provides prescriptive guidance and recommendations that help you quickly address those problems and optimize solutions. When a workload completes, diagnostic information about the job or query and the cluster that processed them is collected by Telemetry Publisher, a role in the Cloudera Manager Management Service, and sent to Cloudera Observability.

Cloudera Observability enables you to interactively understand your environment, data services, workloads, clusters, and resources, and optimize your systems through:
  • A wide range of metrics and health tests that help you identify and troubleshoot both existing and potential issues.
  • Prescriptive guidance and recommendations that help you quickly address those problems and optimize solutions.
  • Performance baselines and historical analysis that help you identify and address performance problems.
In addition, Cloudera Observability also enables you to:
  • Visually display your workload cluster’s current and historical costs that help you plan and forecast budgets, future workload environments, and justify current user groups and resources.
  • Trigger actions in real-time across jobs and queries that help you take steps to alleviate potential problems.
  • Enable the daily delivery of your cluster statistics to your email address that help you to track, compare, and monitor without having to log in to the cluster.
  • Break down your workload metrics into more meaningful views for your business requirements that help you analyze specific workload criteria. For example, you can analyze how queries that access a particular database or that use a specific resource pool are performing against your SLAs. Or you can examine how all the queries are performing on your cluster that are sent by a specific user.