Languages and Compilers for Parallel Computing: 20th International Workshop, LCPC 2007, Urbana, IL, USA, October 11-13, 2007, Revised Selected Papers

ISBN-13: 9783540852605 / Angielski / Miękka / 2008 / 354 str.

Vikram Adve;Maria Jesus Garzaran;Paul Petersen

Opis

It is our pleasure to present the papers from the 20th International Workshop on Languages and Compilers for Parallel Computing For the past 19 years, this workshop has been one of the primary venues for presenting and learning about a wide rangeof currentresearchin parallelcomputing. We believe that tradition has continued in this, the 20th year of the workshop. This year, we received 49 paper submissions from 10 countries. About a quarter of the papers (12 out of 49) included authors from industry. We selected 23 papers to be presented at the workshop, for an acceptancerate of 47%, which was similar to that of the last two years. Each paper received at least three reviews, with about two-thirds of the papers getting four or more reviews each. Most papers also received at least one review from an external reviewer. The committeeheldafull-dayteleconferencetodiscussthereviewsandselectpapers. ProgramCommittee memberswho had a con?ict with a paper left the call when that paper was being discussed. There were seven submissions that included ProgramCommittee members as co-authors. These papers were evaluated more stringently and four of seven were accepted. Theworkshopthis yearalsoincludedtwoexcitingspecialevents.First, David Kirk, Chief Scientist of nVidia and a member of the National Academy of En- neering, gaveakeynotetalkonusinghighlymultithreadedgraphicsprocessorsfor accelerating general-purpose parallel computing applications. Kirk and nVidia have led the drive to make the high parallelism in graphics processors more e- ilyaccessibleforawiderangeofapplicationsbeyondtraditionalgraphicsproce- ing, andthistalkgaveLCPCattendeesavaluableperspectiveonthepotentialof thiswork.

Spis treści

Reliability.- Compiler-Enhanced Incremental Checkpointing.- Techniques for Efficient Software Checking.- Languages.- Revisiting SIMD Programming.- Multidimensional Blocking in UPC.- An Experimental Evaluation of the New OpenMP Tasking Model.- Language Extensions in Support of Compiler Parallelization.- Parallel Compiler Technology I.- Concurrency Analysis for Shared Memory Programs with Textually Unaligned Barriers.- Iteration Disambiguation for Parallelism Identification in Time-Sliced Applications.- A Novel Asynchronous Software Cache Implementation for the Cell-BE Processor.- Pillar: A Parallel Implementation Language.- Associative Parallel Containers in STAPL.- Explicit Dependence Metadata in an Active Visual Effects Library.- Supporting Huge Address Spaces in a Virtual Machine for Java on a Cluster.- Modeling Relations between Inputs and Dynamic Behavior for General Programs.- Evaluation of RDMA Opportunities in an Object-Oriented DSM.- Automatic Communication Performance Debugging in PGAS Languages.- Parallel Compiler Technology II.- Exploiting SIMD Parallelism with the CGiS Compiler Framework.- Critical Block Scheduling: A Thread-Level Parallelizing Mechanism for a Heterogeneous Chip Multiprocessor Architecture.- Languages II.- Capsules: Expressing Composable Computations in a Parallel Programming Model.- Communicating Multiprocessor-Tasks.- An Effective Automated Approach to Specialization of Code.- Flow-Sensitive Loop-Variant Variable Classification in Linear Time.- Using ZBDDs in Points-to Analysis.

Blurb

This book constitutes the thoroughly refereed post-conference proceedings of the 20th International Workshop on Languages and Compilers for Parallel Computing, LCPC 2007, held in Urbana, IL, USA, in October 2007.

The 23 revised full papers presented were carefully reviewed and selected from 49 submissions. The papers are organized in topical sections on reliability, languages, parallel compiler technology, libraries, run-time systems and performance analysis, and general compiler techniques.