Languages and Compilers for Parallel Computing

LCPC'98 Steering and Program Committes for their time and energy in - viewing the submitted papers. Finally, and most importantly, we thank all the authors and participants of the workshop. It is their signi cant research work and their enthusiastic discussions throughout the workshopthat made LCPC'98 a success. May 1999 Siddhartha Chatterjee Program Chair Preface The year 1998 marked the eleventh anniversary of the annual Workshop on Languages and Compilers for Parallel Computing (LCPC), an international - rum for leading research groups to present their current research activities and latest results. The LCPC community is interested in a broad range of te- nologies, with a common goal of developing software systems that enable real applications. Amongthetopicsofinteresttotheworkshoparelanguagefeatures, communication code generation and optimization, communication libraries, d- tributed shared memory libraries, distributed object systems, resource m- agement systems, integration of compiler and runtime systems, irregular and dynamic applications, performance evaluation, and debuggers. LCPC'98 was hosted by the University of North Carolina at Chapel Hill (UNC-CH) on 7 - 9 August 1998, at the William and Ida Friday Center on the UNC-CH campus. Fifty people from the United States, Europe, and Asia attended the workshop. The program committee of LCPC'98, with the help of external reviewers, evaluated the submitted papers. Twenty-four papers were selected for formal presentation at the workshop. Each session was followed by an open panel d- cussion centered on the main topic of the particular session.

Includes supplementary material: sn.pub/extras

Inhalt
Java.- From Flop to MegaFlops: Java for Technical Computing.- Considerations in HPJava Language Design and Implementation.- Locality.- A Loop Transformation Algorithm Based on Explicit Data Layout Representation for Optimizing Locality.- An Integrated Framework for Compiler-Directed Cache Coherence and Data Prefetching.- I/O Granularity Transformations.- Network Computing.- Stampede A Programming System for Emerging Scalable Interactive Multimedia Applications.- Network-Aware Parallel Computing with Remos.- Object-Oriented Implementation of Data-Parallelism on Global Networks.- Fortran.- Optimized Execution of Fortran 90 Array Language on Symmetric Shared-Memory Multiprocessors.- Fortran RED A Retargetable Environment for Automatic Data Layout.- Automatic Parallelization of C by Means of Language Transcription.- Irregular Applications.- Improving Compiler and Run-Time Support for Irregular Reductions Using Local Writes.- Beyond Arrays A Container-Centric Approach for Parallelization of Real-World Symbolic Applications.- SIPR: A New Framework for Generating Efficient Code for Sparse Matrix Computations.- HPF-2 Support for Dynamic Sparse Computations.- Instruction Scheduling.- Integrated Instruction Scheduling and Register Allocation Techniques.- A Spill Code Placement Framework for Code Scheduling.- Copy Elimination for Parallelizing Compilers.- Potpourri.- Compiling for SIMD Within a Register.- Automatic Analysis of Loops to Exploit Operator Parallelism on Reconfigurable Systems.- Principles of Speculative RunTime Parallelization.- Dependence Analysis.- The Advantages of Instance-Wise Reaching Definition Analyses in Array (S)SA.- Dependency Analysis of Recursive Data Structures Using Automatic Groups.- The I+ Test.