Bass Wiki

From Wiki

The Biomedical Analysis and Simulation Supercomputer (BASS, pronounced like base) consists of 452 CPUs tightly coupled to each other and to 180 GPU Computing Processors that function as image and geometry calculation accelerators, providing the equivalent computing power of over thirteen thousand processors for image-intensive applications.

• A summary of the specifications is available on the specifications page.

The BASS was purchased with funding from the National Institutes of Health's National Center for Research Resources, through their High-End Instrumentation program award number NIH 1S10RR023069-01. This award number must appear in the acknowledgments section of all publications that report results obtained using the system. As part of a Strategic Relationship Agreement, Sun provided significant special educational discounts. NVIDIA provided discounts on the CUDA hardware and donated 40 of the GPU computing processors.

• A short description of the purpose for the machine is available on the Purpose page.
• A Thank you! to those who made the system possible is on the acknowledgments page.

Getting started

• Get an account
• Howto page for system users
• Current status and planned downtime/upgrades
• Software Page Installed programs and libraries (and requests for new ones).
• BASS FAQ Frequently-asked questions.

Upcoming Events of Interest

UNC Research Computing Courses: UNC Research Computing is offering the following courses this spring. For more details and to register follow this link:

• 2/9/2010 Using Emerald, a Research Computing Server
• 2/10/2010 Using Topsail, a Research Computing Server
• 2/18/2010 Introduction to Scientific Computing
• 3/2/2010 OpenMP for Parallel Computing
• 4/6/2010 Using Emerald, a Research Computing Server

Tar Heel Grid ITS Research Computing and RENCI are pleased to announce the deployment of the Tar Heel Grid. The Tar Heel Grid is a distributed computing system designed to provide high throughput computing (HTC) services to solve problems that require large amounts of computing cycles in a relatively short time. Information about the Tar Heel Grid can be found here.

Other information

• Administrators' Page
• Testing Page
• Publications Page
• Wish list for new hardware on the machine
• Infrastructure: Information on Power and Data
• External:Links Links to outside stories or pages of potential interest.
• Page for discussion of the computational-science needs and assets on campus