Research Computing

Research computing is the function, commonly an organized department at a university, that manages the HPC clusters and computing resources available to a campus or institution. A research computing group is responsible for the setup and ongoing management of both HPC infrastructure and the HPC applications that researchers depend on, bridging the gap between individual scientists and the complex systems their work requires.

What research computing does

Researchers are usually experts in their own domain, not in operating a supercomputer. Research computing exists to close that gap. Typical responsibilities include:

• Operating shared infrastructure, Provisioning, patching, and maintaining the cluster and its supporting infrastructure so it stays available and secure.
• Managing applications, Installing, updating, and tuning the scientific software and libraries that different research groups need.
• Scheduling and allocation, Running the job scheduler and setting fair-share policies so many groups can share limited resources.
• User support and training, Helping researchers submit jobs, size their workloads, and use the system effectively.

Why research computing matters

By centralizing HPC expertise, research computing lets a whole campus benefit from resources that no single lab could justify or operate on its own. It turns a high performance computing system into a productive shared service, supporting a broad mix of workloads, from genomics to climate modeling to AI, across many departments at once. The result is that domain scientists can focus on their research questions while a dedicated team handles the underlying systems, from the hardware in the HPC data center to the software running on it.