OpenMP: Memory, Devices, and Tasks: 12th International Workshop on OpenMP, IWOMP 2016, Nara, Japan, October 5-7, 2016, Proceedings (Lecture Notes in Computer Science Book 9903)

This book constitutes the proceedings of the 12th International Workshop on OpenMP, IWOMP 2016, held in Nara, Japan, in October 2016. The 24 full papers presented in this volume were carefully reviewed and selected from 28 submissions. They were organized in topical sections named: applications, locality, task parallelism, extensions, tools, accelerator programming, and performance evaluations and optimization. Front Matter....Pages I-XI Front Matter....Pages 1-1 Estimation of Round-off Errors in OpenMP Codes....Pages 3-16 OpenMP Parallelization and Optimization of Graph-Based Machine Learning Algorithms....Pages 17-31 Front Matter....Pages 33-33 Evaluating OpenMP Affinity on the POWER8 Architecture....Pages 35-46 Workstealing and Nested Parallelism in SMP Systems....Pages 47-60 Description, Implementation and Evaluation of an Affinity Clause for Task Directives....Pages 61-73 Front Matter....Pages 75-75 NUMA-Aware Task Performance Analysis....Pages 77-88 OpenMP Extension for Explicit Task Allocation on NUMA Architecture....Pages 89-101 Approaches for Task Affinity in OpenMP....Pages 102-115 Towards Unifying OpenMP Under the Task-Parallel Paradigm....Pages 116-129 A Case for Extending Task Dependencies....Pages 130-140 OpenMP as a High-Level Specification Language for Parallelism....Pages 141-155 Scaling FMM with Data-Driven OpenMP Tasks on Multicore Architectures....Pages 156-170 Front Matter....Pages 171-171 Reducing the Functionality Gap Between Auto-Vectorization and Explicit Vectorization....Pages 173-186 A Proposal to OpenMP for Addressing the CPU Oversubscription Challenge....Pages 187-202 Front Matter....Pages 203-203 Testing Infrastructure for OpenMP Debugging Interface Implementations....Pages 205-216 The Secrets of the Accelerators Unveiled: Tracing Heterogeneous Executions Through OMPT....Pages 217-236 Language-Centric Performance Analysis of OpenMP Programs with Aftermath....Pages 237-250 Front Matter....Pages 251-251 Pragmatic Performance Portability with OpenMP 4.x....Pages 253-267 Multiple Target Task Sharing Support for the OpenMP Accelerator Model....Pages 268-280 Early Experiences Porting Three Applications to OpenMP 4.5....Pages 281-292 Front Matter....Pages 251-251 Design and Preliminary Evaluation of Omni OpenACC Compiler for Massive MIMD Processor PEZY-SC....Pages 293-305 Front Matter....Pages 307-307 Evaluating OpenMP Implementations for Java Using PolyBench....Pages 309-319 Transactional Memory for Algebraic Multigrid Smoothers....Pages 320-335 Supporting Adaptive Privatization Techniques for Irregular Array Reductions in Task-Parallel Programming Models....Pages 336-349 Back Matter....Pages 351-352