Cluster Information
Overview​
The Bayes cluster is intended for running smaller jobs, as well as developing, debugging, and testing codes. It is a self-built and hosted beowulf cluster. Each compute node has 20 cores 2.20 GHz Intel Skylake CPU-cores and 256 GB RAM. There are also two nodes with GPUs (NVIDIA-P100). For more details see the Hardware Configuration section below.
| NodeName | CPU | Memory | GPU | Storage |
|---|---|---|---|---|
| Control | 20 cores | 128GB | NA | 240GB |
| Compute2030005000 | 20 cores | 128GB | NA | 600GB |
| Compute2030005001 | 20 cores | 128GB | NA | 3.0 TB |
| Compute2030005002 | 20 cores | 128GB | Tesla P100 12GB | 4.8TB |
| Compute2030005003 | 20 cores | 128GB | Tesla P100 12GB | 4.8TB |
How to Access the Bayes Cluster​
You have to request an account before using the Bayes cluster, and then log in through SSH.
- Requesting Access to Bayes: if you would like an account on Bayes, please
fill out the Application for to request an accountemail the administrator: bayes [AT] uicstat.com with these information:- Name and Student/Staff ID number
- Which Department
- Propose of Usage
- Consent letter from the supervisor (required for student, research position and visiting staff)
- Logging into Bayes:
ssh <user_name>@hpc.uicstat.com
How to work on the Bayes Cluster​
Since Bayes Cluster is runing a Linux system, knowing some basic Linux commands is hightly recommended. For an introduction to navigating a Linux system, here are some reference:
- Introduction to Linux | by Linux Foundatation
- Introduction to Linux and Command Line | by USC-HPC
Then, please read our guideline of the Bayes cluster to know more about the file systems, module system, and scheduling system running on Bayes cluster.
Notice​
The control node on Bayes should be used for the basic usage (log in, submit your job, file transfer, and etc.) only. No jobs should be run on the control node, other than brief tests that last no more than a few minutes.