Scientific Computing Applications on a Stream Processor

Abstract

Stream processors, developed for the stream programming model, perform well on media applications. In this paper we examine the applicability of a stream processor to scientific computing applications. Eight scientific applications, each having different performance characteristics, are mapped to a stream processor. Due to the novelty of the stream programming model, we show how to map programs in a traditional language, such as FORTRAN. In a stream processor system, the management of system resources is the programmers' responsibility. We present several optimizations, which enable mapped programs to exploit various aspects of the stream processor architecture. Finally, we analyze the performance of the stream processor and the presented optimizations on a set of scientific computing applications. The stream programs are from 1.67 to 32.5 times faster than the corresponding FORTRAN programs on an Itanium 2 processor, with the optimizations playing an important role in realizing the performance improvement.

Accelerate the Training Process of BP neural Network with CUDA Technology

“…Stream processor can be considered as the concrete implementation of stream computing model, as well as NVIDIA GPUs being typical stream processor device. But GPGPU (General-Purpose computing on Graphics Processing) [5,10] greatly limit the powerful parallel processing ability of GPU for too much reliance on graphical API interface. CUDA (Compute Unified Device Architecture) technique provides direct access interface of hardware, and extricates from using graphical API interface to access GPU which is indirect realization of GPGPU.…”

A novel architecture scheme with adaptive pipeline coupling technique for DSP processor design

“…High performance digital signal processing becomes more and more important for the real application in different domain [1] [2]. These applications always need to deal with great amount of data, lots of loop processes.…”