University of Houston Computer Science Webcast
On May 28, 2009, Babara Chapman, Professor of Computer Science at the University of Houston, will present on parallel programming models for multicore platforms.
Webcast Details
Title: Parallel Programming for Multicore Platforms
Date: Thursday, May 28, 2009
Time: 11:00AM - 12:00PM ET
Parallel computing is now a mass movement. As individual processing cores are re-engineered to save power, industry is combining them in order to provide accustomed levels of performance increases. Tomorrow's applications will therefore need to run on multiple cores and possibly many cores.
A major concern for application developers is the need to express the parallelism in their computations in a manner that permits efficient exploitation of the target platform, yet does not require months or years of code redevelopment. Moreover, parallel programs need to offer the same level of portability enjoyed by sequential programs.
MPI is widely used for distributed memory parallel programming. OpenMP is becoming increasingly popular as an application programming interface for shared memory code development. Both MPI and OpenMP may be used together with C or Fortran applications. Additionally, the partitioned global address space (PGAS) languages have been proposed to support the programming of systems that include cores distributed across a network.
In this webcast, Dr. Chapman will:
- Discuss the pros and cons of each of these different language extensions for parallel programming.
- Examine how they can be used to develop applications on systems such as those provided by SiCortex.
- Address OpenMP in more depth.
- Share insights into how OpenMP will evolve to meet the needs of future applications developers.
