Document Type

Technical Report

Department

Computer Science and Engineering

Publication Date

2006-01-01

Filename

wucse-2006-50.pdf

Technical Report Number

WUCSE-2006-50

Abstract

Pipelining an algorithmis a popularmethod of increasing the performance of many computation-intensive applications. Often, one wants to form pipelines composed mostly of commonly used simple building blocks such as DSP components, simple math operations, encryption, or pattern matching stages. Additionally, one may desire to map these processing tasks to different computational resources based on their relative performance attributes (e.g., DSP operations on an FPGA). Auto-Pipe is composed of the X Language, a flexible interface language that aids the description of complex dataflow topologies (including pipelines); X-Com, a compiler for the X Language; X-Sim, a tool for modeling pipelined architectures based on measured, simulated, or derived task and communications behavior; X-Opt, a tool to optimize X applications under various metrics; and X-Dep, a tool for the automatic deployment of X-Com- or X-Sim-generated applications to real or simulated devices. This thesis presents an overview of the Auto-Pipe system, the design and use of the X Language, and an implementation of X-Com. Applications developed using the X Language are presented which demonstrate the effectiveness of describing algorithms using X, and the effectiveness of the Auto-Pipe development flow in analyzing and improving the performance of an application.

Comments

Permanent URL: http://dx.doi.org/10.7936/K7J964NR

Share

COinS