Why Telemetry Framework?¶
In the modern datacenter, things are dynamic. Infrastructure needs to be monitored at a high polling interval, and actions taken quickly. Our environments are dynamic, constantly changing in the cloud era. An ability to monitor and react to these changing conditions with a strong degree of flexability is why the telemetry framework was created.
The telemetry framework is itself a dynamic application running atop OpenShift (Kubernetes) using several components such as Prometheus, the Smart Gateway, collectd and the Apache QPID Dispatch Router. With a strong desire for low latency in message delivery and high resolution information, the telemetry framework is an excellent core to building the information to automatically react to situations in your cloud infrastructure.
The telemetry framework gives you the core functionality to start gathering data, storing it, and then reacting to it. With Prometheus Alertmanager you can build rules that will create metrics-driven events that allow for notifications to various endpoints. The flexibility of the Prometheus time-series database can allow for predicting imminent failures before they happen.
You can add other systems as well to provide things like graphing, data correlation, traps, and automation.