High Performance Computing
USC's Research Computing has multiple High Performance Computers configured for ultra fast calculations of research based scientific inquiry and formulas. Whether predicting weather patterns, analyzing soil samples, or testing molecular interactions, the Research Computing Center has multiprocessor clusters ready to deliver fast results.
Research Computing Center provides the USC community with access to high performance computing systems, remote and collaborative visualization resources, data storage, user services & software packages, assistance with HPC code development, and opportunities for computational research collaborations. The primary funding for hardware was made possible by the sharing of resources and cooperation among groups of researchers, along with substantial investments by NSF and the College of Arts and Sciences in the EPSCoR RII Track I and Track II programs. NSF EPSCoR RII and substantial cost sharing by the university provide personnel resources.
USC HPC Cluster User Guides
RCC resources are configured to serve a large and diverse user research community of the University of South Carolina. In order to keep those resources functional and in good shape, the users must be aware of the RCC Usage Policies. Failure to comply with the Usage Policies may result in the suspension and closure of the user accounts.
Contact the Research and Scientific Computing support personel for assistance with account creation or other High Performance Computing needs.
RCC staff has created documentation related to the Usage Policies guidelines and made them available on the web, as well as on the RCC User Portal. The users should be familiarized with the Usage Policies and assume responsibility, especially for the issues related to the security and data management.
Any Computing Account should be used only by the User for which that account was created. RCC resources should also be used for the documented or intended purposes. The following activities examples of the improper use of the RCC resources:
- Using HPC resources to run applications that can be run on a desktop
- Running jobs on the head and login nodes on the HPC resources
- Using storage systems for backing up data that is not related to research activity
It is very important to maintain the security of the RCC computing resources. Here are few guidelines to be insured by PI’s:
- Information such as login name and passwords must never be shared with anyone.
- PIs should vouch for and request allocation for their students, postdocs, etc
- PIs should notify the RCC about the closure of an account when projects are completed and student accounts are no longer needed.
The users of the RCC computing resources are encouraged to request assistance from RCC staff (COCASRrcchelp@cas.sc.edu). RCC support staff can provide advice with the development, debugging, porting, and optimizing the codes. However, such support has somewhat limited extent to actual code development, porting, debugging, and optimization.
Your acknowledgement of the RCC computing and human resources is necessary and very much appreciated. Please reference RCC in any publications or research reports. Your help is important for reporting activities, maintaining the systems, and acquiring new resources. We suggest to use the following acknowledgement as a guideline:
The authors acknowledge the Research Computing Center at the University of South Carolina for providing the computing resources that contributed to the results of this paper.
Allocations on the RCC resources are available to the all community of the USC. However, higher priority is given to the researches and students that are closely involved into the EPSCoR related projects (NSF EPSCoR desktop to petascale award funds).
The RCC is generally allocated to a single PI which can be a faculty or a research staff at the USC. The PIs can request allocation for their students and postdoctoral researchers and assume responsibility for the activity of such accounts. The PIs should make sure that the resources are being used in accordance with the RCC Usage Policies.
When applying for an allocation on the high performance computing (HPC) systems, you will be given an account on one or more RCC clusters. We do not assign a specific number of computing hours (units) to a particular HPC allocation but rather limit number of processors a researcher can use simultaneously on a HPC system. Currently the maximum number of processors is set to 60. Such limitation does not apply to allocations that are related to the EPSCoR funded projects.