Skip to main content


eCommons@Cornell

eCommons@Cornell >
College of Engineering >
Computer Science >
Computer Science Technical Reports >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1813/6713
Title: Perfect Pipelining: A New Loop Parallelization Technique
Authors: Aiken, Alexander
Nicolau, Alexandru
Keywords: computer science
technical report
Issue Date: Oct-1987
Publisher: Cornell University
Citation: http://techreports.library.cornell.edu:8081/Dienst/UI/1.0/Display/cul.cs/TR87-873
Abstract: Parallelizing compilers do not handle loops in a satisfactory manner. Fine-grain transformations capture irregular parallelism inside a loop body not amenable to coarser approaches but have limited ability to exploit parallelism across iterations. Coarser methods sacrifice irregular forms of parallelism in favor of pipelining (overlapping) iterations. In this paper we present a new transformation, Perfect Pipelining, that bridges the gap between these fine-and coarse-grain transformations while retaining the desirable features of both. This is accomplished even in the presence of conditional branches and resource constraints. For loops typically encountered in practice, Perfect Pipelining achieves the effect of full loop unrolling coupled with fine-grain parallelization. To make our claims rigorous, we develop a formalism for parallelization. The formalism can also be used to compare transformations across computational models. As an illustration, we show that Doacross, a transformation intended for synchronous and asynchronous multiprocessors, can be expressed as a restriction of Perfect Pipelining.
URI: http://hdl.handle.net/1813/6713
Appears in Collections:Computer Science Technical Reports

Files in This Item:

File Description SizeFormat
87-873.pdf2.57 MBAdobe PDFView/Open
87-873.ps749.53 kBPostscriptView/Open

Refworks Export

Items in eCommons are protected by copyright, with all rights reserved, unless otherwise indicated.

 

© 2014 Cornell University Library Contact Us