Princeton Research Computing provides various high-performance computing clusters in support of Princeton's Research and Teaching Mission. Systems range from small, entry-level servers to large scale CPU/ GPU clusters.

Research centric professional services in data storage & management, visualization, secured data enclaves, and software engineering enables our scientific community to pursue new ideas in every academic field.

For more information please visit the Research Computing website.

Research Computing Clusters

For details of available computing clusters summarized below, please visit Research Computing Systems.

Small Systems

• Nobel and Adroit are small systems used for classwork and program development.

Large Clusters

• Tiger is the newest, most powerful system.  It includes 320 Nvidia P100 GPUs along with 392 non-GPU nodes.
• Tigressdata is used for post-processing and visualization of the results from jobs run on the large clusters.
• Della is a large general-purpose cluster.  It is used for both parallel and serial jobs.
• Perseus is a large cluster that is used for large, computationally intensive non-GPU jobs.
• Traverse is composed of 46 IBM POWER9 nodes with four NVIDIA V100 GPUs per node. The cluster is primarily intended to support research at the Princeton Plasme Physics Lab (PPPL).

Research Computing Software

Faculty, Researchers, and students can access statistical applications, design tools, analysis, and visualization software on all available clusters.

For additional details on software readily available on the computing clusters, please visit Research Computing Software.