Monitoring Nutanix Prism

Figure 1 depicts the specialized web-based monitoring model that eG Enterprise provides for the Nutanix Prism.

Figure 1 : Layer model of the Nutanix Prism

Each layer of Figure 1 is mapped to tests that invoke REST API commands to report a wide variety of metrics related to Prism health and the health of the cluster(s) managed by the Prism. Using these metrics, administrators can find quick and accurate answers to the following performance queries:

  • Is the Prism available? If so, how quickly is it responding to HTTP requests?
  • Which clusters are being managed by the Prism?
  • Is any cluster over-utilizing its compute and storage resources?
  • How is the I/O load on containers and/or storage pools? Is any container/storage pool overloaded with I/O requests?
  • Is any container/storage pool utilizing bandwidth excessively when serving I/O requests?
  • Is any container/storage pool latent?
  • Is any container/storage pool running out of storage space?

This topic elaborates on each of the top 3 layers of Figure 1.