Description

Ganglia is an open-source, scalable, distributed monitoring system for high-performance computing systems such as clusters and grids. It is designed to monitor and visualize the vital metrics of a system such as the usage of the CPU, memory, disk and network resources in real-time. It also provides alerts when any of these resources exceed pre-defined thresholds. Ganglia is composed of a core monitoring daemon called gmond and a graphical web interface called Ganglia Web. Gmond collects metrics from the local system and broadcasts them to other gmond instances in the cluster or grid. It is scalable since it can be deployed to monitor thousands of servers, and has a low overhead. Ganglia Web is an easy-to-use web interface that provides an intuitive way to view and analyze the metrics collected by gmond. It displays the metrics in an interactive, graphical manner and allows for customization of the graphs for comparison of multiple systems. It also allows for drill-down analysis of the metrics to pinpoint system bottlenecks.

Categories