This monitor tracks the memory utilization on the head node. The memory utilization is based on the corresponding Windows performance counter.

High memory usage on the head node can be caused by HPC-related processes, or by other services, applications, or processes that are running on the head node.

• Use perfmon or other administrative tools to analyze the memory usage of processes on the head node. If you are running services on the head node that are not required for the management or administration of the HPC cluster, consider moving those services to other computers. If you have a large number of cluster nodes and the high memory usage is due to SQL Server Express on the head node, consider installing the HPC databases on one or more remote servers that are running SQL Server. If necessary, back up the head node and reinstall Microsoft HPC Pack on a computer that has greater memory capacity.

• You can optionally add a recovery task in Operations Manager to automatically trigger custom actions if the memory utilization exceeds the configured threshold.