D
David Padua
Researcher at University of Illinois at Urbana–Champaign
Publications - 244
Citations - 11510
David Padua is an academic researcher from University of Illinois at Urbana–Champaign. The author has contributed to research in topics: Compiler & Automatic parallelization. The author has an hindex of 58, co-authored 243 publications receiving 11155 citations. Previous affiliations of David Padua include New Jersey Institute of Technology & University of Málaga.
Papers
More filters
Advanced Program Restructuring for High-Performance Computers with Polaris
William Blume,Ramón Doallo,Rudolf Eigenmann,John Grout,Jay Hoeflinger,Thomas R. Lawrence,Jaejin Lee,David Padua,Yunheung Paek,Bill Pottenger,Lawrence Rauchwerger,Peng Tu +11 more
TL;DR: Recent experience at Illinois on the automatic parallelization of scientific codes using the Polaris restructurer is discussed and several new analysis techniques that have been developed in recent years based on an extensive analysis of the characteristics of real Fortran codes are presented.
Journal ArticleDOI
Static and dynamic evaluation of data dependence analysis techniques
Paul Petersen,David Padua +1 more
TL;DR: The results indicate that Banerjee's test is for all practical purposes as accurate as the more complex Omega test in detecting parallelism, but the Omega test is quite effective in proving the existence of dependences, in contrast with Baners' test, which can only disprove, or break dependences.
Proceedings ArticleDOI
The cedar system and an initial performance study
David Kuck,Edward S. Davidson,Duncan H. Lawrie,Ahmed H. Sameh,Chuan-Qi Zhu,Alexander V. Veidenbaum,J. Konicek,Pen-Chung Yew,Kyle A. Gallivan,William Jalby,Harry A. G. Wijshoff,R. Bramley,Ulrike Meier Yang,Perry A. Emrath,David Padua,Rudolf Eigenmann,Jay Hoeflinger,Greg P. Jaxon,Zhiyuan Li,T. Murphy,J. Andrews +20 more
TL;DR: An overview of the Cedar multiprocessor is given and a methodology for judging parallel system performance is presented that applies to Cedar, Cray YMP-8, and Thinking Machines CM-5.
Journal ArticleDOI
Utilizing multidimensional loop parallelism on large scale parallel processor systems
TL;DR: Optimal processor assignment algorithms are presented for simple and complex nested parallel loops that result in almost linear speedups on parallel processor machines with a few tens of processes and significantly high speedups for machines with hundreds or thousands of processors.
Journal ArticleDOI
In search of a program generator to implement generic transformations for high-performance computing
Albert Cohen,Sébastien Donadio,Maria-Jesus Garzaran,Christoph Herrmann,Oleg Kiselyov,David Padua +5 more
TL;DR: In this paper, the authors explore in-between solutions, neither fully automatic nor fully manual ways to adapt a computationally intensive application to the target architecture, by mimicking complex sequences of transformations useful to optimize real code.