Task and instruction scheduling in parallel multithreaded processors /

A new prioritizatioin, technique based on a critical path

Bibliographic Details
Main Author: Mishra, Amitabh, 1971-
Format: Thesis eBook
Language:English
Published: [Place of publication not identified] : [publisher not identified] ; 1996.
Subjects:
Online Access:Link to OAKTrust copy
Description
Summary:A new prioritizatioin, technique based on a critical path
and multithreaded architectures: the multiple-instruction-
applications chosen carefully to reflect diverse program
based analysis of program graphs is shown to enhance
based on considerations other than a critical path analysis
behavior. Our results suggest the use Of superscalar
enhances instruction throughput in a processor by combining
from the fetched threads to the multiple functional units of
hiding feature of multithreaded architectures. In such
instruction throughput significantly by scheduling the best
instruction throughput UP to 20% better than previously
issue ability of a superscalar processor, and the latency-
multithreaded processors in order to Perform efficient
of the program graph. Our simulations employ Parallel
Parallel muitithreading is a technique to execute parallel
Parallel processing.
processor Pipelines, and by issuing the best instructions
processors, several threads issue instructions to a
program parallelism with the strong features of superscalar
programs on a multithreaded superscalar processor. It
proposed prioritization techniques that employ heuristics
superscalar processors multiple functional units every cycle.
task and instruction scheduling techniques Yield an
the processor. Simulation-based comparisons show that our
threads (or tasks) from a parallel Program on to the
Item Description:"Major subject: Computer Science".
Vita.
Physical Description:ix, 60 leaves : illustrations ; 28 cm.
Also available online.
Issued also on microfiche from Lange Micrographics.
Bibliography:Includes bibliographical references: pages 57-59.