• Publications
  • Influence
Optimizing Compilers for Modern Architectures: A Dependence-based Approach
TLDR
Data dependence is a fundamental compiler analysis tool for optimizing programs on high-performance microprocessors and parallel architectures. Expand
  • 1,077
  • 102
Automatic translation of FORTRAN programs to vector form
TLDR
The theoretical background is developed here for employing data dependence to convert FORTRAN programs to parallel form. Expand
  • 813
  • 43
  • PDF
Compiling Fortran D for MIMD distributed-memory machines
TLDR
We propose to solve this problem by developing the compiler technology needed to establish a machine-independent programming model. Expand
  • 563
  • 27
Conversion of control dependence to data dependence
TLDR
This paper presents a method for systematically converting all control dependences to data dependences by eliminating goto statements and introducing logical variables to control the execution of statements in the program. Expand
  • 612
  • 24
  • PDF
Practical dependence testing
TLDR
This paper proposes a dependence testing scheme based on classifyingpairs ofsubscripted variable references, as well as empirical results showing that these references dominate scientific Fortran codes. Expand
  • 281
  • 16
  • PDF
An Implementation of Interprocedural Bounded Regular Section Analysis
TLDR
Regular section analysis, which summarizes interprocedural side effects on subarrays in a form useful to dependence analysis, while avoiding the complexity of prior solutions, is shown to be a practical addition to a production compiler. Expand
  • 295
  • 16
Automatic data layout for distributed-memory machines
TLDR
The goal of languages like Fortran D or High Performance Fortran is to provide a simple yet efficient machine-independent parallel programming model. Expand
  • 142
  • 15
Maximizing Loop Parallelism and Improving Data Locality via Loop Fusion and Distribution
TLDR
Loop fusion is a program transformation that merges multiple loops into one. Expand
  • 275
  • 13
  • PDF
The rise and fall of High Performance Fortran: an historical object lesson
TLDR
High Performance Fortran failed to achieve the success we all hoped for it. Expand
  • 127
  • 13
Improving cache performance in dynamic applications through data and computation reorganization at run time
TLDR
In this paper, we explore ways to achieve better data reuse in irregular and dynamic applications by building on the inspector-executor method used by Saltz for run-time parallelization. Expand
  • 226
  • 12
  • PDF
...
1
2
3
4
5
...