HPC administrators sometimes assume that if all nodes are functioning, the system is fine. However, the most common issue users have is poor or unexpected application performance. In this case, you need a simple tool to help you understand what’s happening on the nodes.