Stella

Advanced Computational Research Cluster

Getting Access

To request access to this resource, please email hpcteam@temple.edu with a brief explanation of your intended use. If you are a student or postdoctoral researcher, be sure to include the name of your supervisor in your request.

This page is under development!

Stella is a cluster available to selected users within the College of Science and Technology. Assembled in 2024, it was funded through an NSF MRI grant (Major Research Instrumentation Program)

The cluster comprises two types of compute nodes optimized for diverse workloads. The first set includes 14 nodes of DELL PowerEdge R6525 and Lenovo ThinkSystem SR645, each equipped with dual AMD EPYC 7543 processors (64 cores, 2.8 GHz base clock) and 512 GB of RAM. These nodes feature Infiniband HDR interconnect for high-speed communication.

The second set consists of 6 GPU-accelerated nodes based on DELL PowerEdge R7525, each powered by dual AMD EPYC 7343 processors (32 cores, 3.2 GHz base clock) and 256 GB of RAM. These nodes are equipped with three NVIDIA A100 GPUs (PCIe), providing 3x 8192 CUDA cores, 3x 512 Tensor cores, and 3x 80 GB of HBM2e memory for different workloads. They also utilize Infiniband HDR interconnect.

In total, Stella features 1,088 CPU cores, 8 TB of RAM, and 18 NVIDIA A100 GPUs. All of this is supported by an 800TB Dell PowerVault ME4048 storage system, providing shared access to user data across the entire cluster.