Cluster Monitoring

This section covers the monitoring of running clusters.

A very important aspect while working with clusters is monitoring its health. Ankush provides an exhaustive graphical representation and reporting system for variety of aspects.

Monitoring performance graphs at cluster level signifies collective performance of all the nodes of a cluster. Monitoring at cluster level provides the user with an extensive platform to closely monitor cluster utilization levels, Heat maps of nodes based on CPU / memory utilization, metrics based on CPU, memory, network, load and packet etc.

In a nutshell, Monitoring is all about:

• Tiles for Alerts, Warning, Services status, Node down and information
• Cluster & Node-wise System Metrics: CPU, Memory, Network, Load and Packet
• Mails for Alerts, Warning, Services status, Node down

A sample screenshot is as shown below:

![] (https://raw.githubusercontent.com/wiki/impetus-opensource/ankush/images/Image90.png) Figure 90: Cluster Monitoring - Generic features