Windows HPC Server 2008 Monitoring and Management
< Go Back
Cluster administrators have said that they want to be able to easily manage their large-scale clusters with heterogeneous compute nodes. They want to be able to view cluster status at a glance, to be able to quickly identify deviance, and to have the ability to drill down for more information. They have said that they want the ability to correlate node and job information and to track changes. They also want to be able to integrate their monitoring and managing functions with their existing infrastructure. Windows HPC Server 2008 provides all of this functionality, and much more.
This video provides a detailed description of all aspects of monitoring and managing in Windows HPC Server 2008, including an in-depth discussion of the integrated view and pivoting capabilities of the Admin Console. The presenter demonstrates how to use the various monitoring and managing features, such as searching in the Nodes List View, filtering (including the creation of custom groups), and sorting. The presenter shows how you can open a remote desktop session to one or more of the compute nodes to perform operations. Then, see a demonstration of the monitoring heat map in the Admin Console, a useful function for monitoring large-scale clusters. Finally, the operations history view is explained and demonstrated.
