The stream architecture is a novel microprocessor architecture with wide application potential. It is critical to study how to use the stream architecture to accelerate scientific computing programs. However, existing stream processors and stream programming languages are not designed for scientific computing. To address this issue, we design and implement a 64-bit stream processor, Fei Teng 64 (FT64), which has a peak performance of 16 Gflops. FT64 supports two kinds of communications, message passing and stream communications, based on which, an interconnection architecture is designed for a FT64-based high-performance computer. This high-performance computer contains multiple modules, with each module containing eight FT64s. We also design a novel stream programming language, stream Fortran 95 (SF95), together with the compiler SF95 compiler, so as to facilitate the development of scientific applications. We test nine typical scientific application kernels on our FT64 platform to evaluate this design. The results demonstrate the effectiveness and efficiency of FT64 and its compiler for scientific computing.

In this paper, we present a detailed case study of the optimizing implementation of a fundamental scientific kernel, matrix-vector multiplication, on FT64, which is the first 64-bit stream processor designed for scientific computing. The major novelties of our study are as follows. First, we develop four stream programs according to different stream organizations, involving dot product, row product, multi-dot product and multi-row product approaches. Second the optimal strip size for partitioning the large matrix is put forward based on a practical parameter model. Finally loop unrolling and software pipelining are used to hide the communications with the computations. The experimental results show that the optimizing implementations on FT64 achieve high speedup over the corresponding Fortran programs running on Itanium 2. It is certain that matrix-vector multiplication can efficiently exploit the tremendous potential of FT64 stream processor through programming optimizations.

The efficiency of scientific applications on the Imagine stream processor is increasingly concerned by researchers. One of the obstacles is that the programming language of Imagine does not target the scientific computing. This paper introduces a program transformation algorithm to automatically transform loops to the stream programs executed on Imagine. The optimization for memory accessing is also considered during the transformation. We have implemented the transformation and optimization algorithm with the GFORTRAN frontend. Preliminary results over benchmark kernels show that our approach is a convenient and efficient solution to develop scientific applications on the Imagine stream processor.

The fast and accurate evaluation of transcendental functions (e.g. exp, log, sin, and atan) is quite important in many domains. We implement a software inline function library that can be called from KernelC programming language to compute 8 typical functions on Imagine architecture. By exploiting some of the key features of Imagine architecture, we have been able to provide single precision transcendental functions that are very accurate yet can typically be evaluated to get 16 function values in between 18 and 43 clock cycles. In this paper, we also discuss the algorithms and implementation details of these functions.