Author

Marc Torrant

Abstract

Many enhancements have been made to the traditional general purpose load-store computer architectures. Among the enhancements are memory hierarchy improvements, branch prediction, and multiple issue processors. A major problem that exists with current microprocessor design is the disparity in the much larger increase in speed of the CPU versus the moderate increase in speed accessing main memory. The simultaneous multithreaded architecture is an extension of the single-threaded architecture that helps hide the performance penalty created by long-latency instructions, branch mispredictions, and memory accesses. Simultaneous multithreaded architectures use a more flexible parallelism, which takes advantage of both instruction-level, and thread-level parallelism. The goal of this project was to design, simulate, and analyze a model of a simultaneous multithreaded architecture in order to evaluate design alternatives. The simulator was created by modifying a version of the Simple Scalar toolset, developed at the University of Wisconsin. The simulations provide documentation for an overall system performance improvement of a simulta neous multithreaded architecture. In early simulation results, performed with the same number of functional units, an improvement in the number of instructions per cycle (IPC) of between 43% and 58% was found using four threads versus a single thread. The horizontal waste rate, which measures the number of unused issue slots, was reduced between 35% and 46%. The vertical waste rate, which measures the percentage- of unused issue cycles (no issue slots used in a cycle), was reduced between 46% and 61%. These results are derived from a set of four sample programs. It was also found that increasing the number of certain functional units did not improve performance, whereas increasing the number of other types of functional units did have a significant positive impact on performance.

Library of Congress Subject Headings

Microprocessors--Design and construction; Memory management; Computer architecture

Publication Date

8-1-1999

Document Type

Thesis

Department, Program, or Center

Computer Engineering (KGCOE)

Advisor

Czernikowski, Roy

Advisor/Committee Member

Hsu, Ken

Comments

Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works. Physical copy available through RIT's The Wallace Library at: TK7895.M5 T67 1999

Campus

RIT – Main Campus

Share

COinS