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.

In the continuing story of monitoring HPC systems, we look at code that measures process, network, and disk metrics.

One goal of HPC administration is effective monitoring of clusters. In this article, we talk about writing code that measures processor and memory metrics on each node.

In rapidly growing HPC installations, you need to understand what is happening within the system to make improvements or simply to justify the purchase.

The Xeon Phi accelerator card from Intel takes an unusual approach: Instead of GPUs, the Xeon Phi features a cluster of CPUs for easier programming.

Introduction of the new OpenMP specification abstracts away many of the thorny issues associated with today’s HPC hardware.

Even with tons of cores per node today, the traditional sets of tools are still serial-only, utilizing a single core; however, some of the more popular tools have parallel versions, allowing you to use the extra cores either to run the same command in parallel or to perform the same task across multiple nodes.

The language you choose to use affects I/O patterns and performance. We track a simple write I/O pattern with C and look at how to improve performance.

Hadoop version 2 expands Hadoop beyond MapReduce and opens the door to MPI applications operating on large parallel data stores.

Your parallel application is running fine, but you want it to run faster. Naturally, you use more and more cores, and everything is great; however, suddenly performance starts decreasing. What just happened?