R
Rudolf Eigenmann
Researcher at University of Delaware
Publications - 187
Citations - 6116
Rudolf Eigenmann is an academic researcher from University of Delaware. The author has contributed to research in topics: Compiler & Automatic parallelization. The author has an hindex of 40, co-authored 183 publications receiving 5914 citations. Previous affiliations of Rudolf Eigenmann include Lawrence Livermore National Laboratory & Purdue University.
Papers
More filters
Journal ArticleDOI
The Cetus Source-to-Source Compiler Infrastructure: Overview and Evaluation
Hansang Bae,Dheya Mustafa,Jae-Woo Lee,Aurangzeb,Hao Lin,Chirag Dave,Rudolf Eigenmann,Samuel P. Midkiff +7 more
TL;DR: An overview and an evaluation of the Cetus source-to-source compiler infrastructure and several techniques that support dynamic optimization decisions are discussed and evaluated.
Book ChapterDOI
Towards OpenMP Execution on Software Distributed Shared Memory Systems
TL;DR: Detailed measurements of the performance characteristics of realistic OpenMP applications from the SPEC OMP2001 benchmarks are presented and application and system characteristics that impede the efficient execution of these programs on a Software DSM system are discussed.
Proceedings ArticleDOI
An Overview of Symbolic Analysis Techniques Needed for the Effective Parallelization of the Perfect Benchmarks
William Blume,Rudolf Eigenmann +1 more
TL;DR: The techniques include: symbolic data dependence tests for nonlinear expressions, constraint propagation, array summary information, and run time tests that will improve the effectiveness of parallelizing Fortran compilers.
Proceedings ArticleDOI
Fast, automatic, procedure-level performance tuning
Zhelong Pan,Rudolf Eigenmann +1 more
TL;DR: An automated performance tuning solution, which partitions a program into a number of tuning sections and finds the best combination of compiler options for each section, and improves the performance of SPEC CPU2000 FP benchmarks by 12% on average over GCC O3, the highest optimization level.
Proceedings ArticleDOI
Reducing parallel overheads through dynamic serialization
Michael Voss,Rudolf Eigenmann +1 more
TL;DR: This work proposes a framework, based on an inspector-executor model, for identifying loops that are dominated by parallel overheads and dynamically serializing these loops, and implements this framework in the Polaris parallelizing compiler and evaluates two portable methods for classifying loops as profitable or unprofitable.