Overview of Scholar
Scholar is a small computer cluster, suitable for classroom learning about high performance computing (HPC). It consists of 7 interactive login servers and 27 batch worker nodes, including 4 with GPUs.
It can be accessed as a typical cluster, with a job scheduler distributing batch jobs onto its worker nodes, or as an interactive resource, with software packages available through a desktop-like environment on its login servers.
If you have a class that you think will benefit from the use of Scholar, you can schedule it for your class through our web page at: Class Account Request. You only need to register your class itself. All students who register for the class will automatically get login privileges to the Scholar cluster. As a batch resource, the cluster has access to typical HPC software packages and tool chains; as an interactive resource, Scholar provides a Linux remote desktop, or a Jupyter notebook server, or an R Studio server. Jupyter and R Studio can be used by students without any reliance on Linux knowledge or experience.
Scholar CPU compute nodes have 20 processor cores, and 64 GB RAM. GPU nodes have 16 processor cores, 192 GB RAM and 1 Tesla V100 32GB GPU. The network fabric for all is 56 Gbps Infiniband.
|Front-Ends||Number of Nodes||Processors per Node||Cores per Node||Memory per Node||Retires in|
|No GPU||4||Two Haswell CPUs @ 2.60GHz||20||512 GB||2023|
|With GPU||3||Two Sky Lake CPUs @ 2.60GHz with one NVIDIA Tesla V100||20||768 GB||2023|
|Sub-Cluster||Number of Nodes||Processors per Node||Cores per Node||Memory per Node||Retires in|
|A||24||Two Haswell CPUs @ 2.60GHz||20||64 GB||2023|
|G||4||Two Skylake CPUs @ 2.10GHz with one NVIDIA Tesla V100 32GB||16||192 GB||2023|
Faculty who would like to know more about Scholar, please read the Faculty Guide