High-Performance Computing on CUNY’s Cray XE6

The CUNY High-Performance Computing Center is excited to offer early access to its new 1,280 CPU Cray XE6 supercomputer to those attending its first workshop on the Unified Parallel C (UPC) and Co-Array Fortran (CAF) on Friday, Mar. 11 in Building 1N, Room 005 from 9:00am to 5:00pm. 

Through simple modifications to C and Fortran, UPC and CAF allow application developers to write parallel code without using the messagepassing interface (MPI) library. Easier to program than traditional approaches with MPI, UPC and CAF support fine-grain parallelism and provides compiler-based debugging for the entire application. This  class will introduce the features of these parallel language extensions, and provide hands-on training with exercises for the new Cray XE6.

Topic that will be covered include:

-Overview and comparison to MPI
-Introduction to the syntax and semantics of UPC and CAF
-Examples, including collectives, linear algebra algorithms for operations, and halo-update techniques for partial differential equations.
-Hands-on exercises

To register for a seat for this class, please fill out an application.

Leave a Reply