Monitoring the YARN services
When we talk about handling big data and multi node clusters for distributed processing, we consider performance and efficiency as major factors. Monitoring of the YARN services includes collection of cluster, node, and service level metrics. Each of the YARN services exposes its monitoring information as JMX MBean object. As a cluster administrator, a person needs to monitor these metrics through detailed graphs and reporting tools, such as Jconsole, Ganglia, and so on. In this section, we'll discuss the different techniques used to monitor the YARN services.
JMX monitoring
JMX are the Java tools used for monitoring and managing applications, objects, and so on. The resources are represented as Managed Bean or simply MBean objects. An MBean represents a resource running in a Java Virtual Machine. The statistical information collected from these resources regarding performance, system resource usage, application events, and such, could be used to fine tune...
