Accelerator: A data-parallel DSL for programming heterogonous systems
- ๐ค Speaker: Satnam Singh, Microsoft Research
- ๐ Date & Time: Wednesday 13 January 2010, 16:30 - 17:30
- ๐ Venue: SS03, Computer Laboratory, William Gates Building
Abstract
This presentation introduces Microsoftโs Accelerator system which is a library that provides data-parallel arrays and operations which can be implemented on GPUs, x64 SIMD multicores and FPG As. The library can be used from C++ or any .NET language e.g. C# or F#. We make the case that for certain kinds of computations this system provides a way to write a computation once and then have it instantiated on wildly different kinds of computing resources. We will also highlight differences between Accelerator and other technologies e.g. CUDA , OpenCL and Ct.
Series This talk is part of the Computer Laboratory Computer Architecture Group Meeting series.
Included in Lists
- All Talks (aka the CURE list)
- bld31
- Cambridge talks
- Computer Laboratory Computer Architecture Group Meeting
- Department of Computer Science and Technology talks and seminars
- Interested Talks
- School of Technology
- SS03, Computer Laboratory, William Gates Building
- Trust & Technology Initiative - interesting events
- yk449
Note: Ex-directory lists are not shown.
![[Talks.cam]](/static/images/talkslogosmall.gif)

Satnam Singh, Microsoft Research
Wednesday 13 January 2010, 16:30-17:30