1. Introduction to Ganglia
Ganglia is a scalable distributed monitoring system designed for HPC (High Performance Computing) clusters. It can monitor and display various status information of nodes in the cluster. It is monitored by the gmond daemon running on each node. Collect data on CPU, memory, hard disk utilization, I/O load, network traffic, etc., and then summarize it under the gmetad daemon process, use rrdtool to store the data, and finally present the historical data in a curve through the PHP page.
The features of Ganglia are as follows:
- Good scalability, layered architecture design can adapt to the needs of large-scale server clusters;
- Low load overhead and support high concurrency;
- Widely support various operating systems (UNIX, etc.) and CPU architectures, and support virtual machines;
2. Composition of Ganglia
The Ganglia monitoring system consists of three parts, namely gmond, gmetad, and webfrontend, whose functions are as follows:
- gmond: Ganglia monitoring daemon is a daemon process that runs on each node that needs to be monitored. It is used to collect the information of this node and send it to other nodes. It also receives the data sent by other nodes. The default monitoring The port is 8649.
- gmetad: Ganglia meta daemon is a daemon process that runs on a data aggregation node, periodically checks the gmond process of each monitoring node and obtains data from there, and then stores the data indicators in the local RRD storage engine.
- webfrontend: It is a web-based graphical monitoring interface. It needs to be installed on the same node as Gmetad. It fetches data from gmetad, reads the RRD database, and generates charts through rrdtool for front-end display. The interface is beautiful, rich, and functional powerful.