Recent Advances in the Message Passing Interface: 19th European MPI Users’ Group Meeting, EuroMPI 2012, Vienna, Austria, September 23-26, 2012. Proceedings

Front Matter....Pages -
MPI 3 and Beyond: Why MPI Is Successful and What Challenges It Faces....Pages 1-9
MPI Functions and Their Impact on Interconnect Hardware....Pages 10-10
The MPI Communication Library for the K Computer: Its Design and Implementation....Pages 11-11
MPI and Compiler Technology: A Love-Hate Relationship....Pages 12-13
Advanced MPI Including New MPI-3 Features....Pages 14-14
Hands-on Practical Hybrid Parallel Application Performance Engineering....Pages 15-15
Adaptive Strategy for One-Sided Communication in MPICH2....Pages 16-26
A Low Impact Flow Control Implementation for Offload Communication Interfaces....Pages 27-36
Improving MPI Communication Overlap with Collaborative Polling....Pages 37-46
Delegation-Based MPI Communications for a Hybrid Parallel Computer with Many-Core Architecture....Pages 47-56
Efficient Multithreaded Context ID Allocation in MPI....Pages 57-66
Collectives on Two-Tier Direct Networks....Pages 67-77
Exploiting Atomic Operations for Barrier on Cray XE/XK Systems....Pages 78-88
Exact Dependence Analysis for Increased Communication Overlap....Pages 89-99
 mpicroscope : Towards an MPI Benchmark Tool for Performance Guideline Verification....Pages 100-109
OMB-GPU: A Micro-Benchmark Suite for Evaluating MPI Libraries on GPU Clusters....Pages 110-120
Micro-applications for Communication Data Access Patterns and MPI Datatypes....Pages 121-131
Leveraging MPI’s One-Sided Communication Interface for Shared-Memory Programming....Pages 132-141
Wait-Free Message Passing Protocol for Non-coherent Shared Memory Architectures....Pages 142-152
An Efficient Kernel-Level Blocking MPI Implementation....Pages 153-162
Automatic Resource-Centric Process Migration for MPI....Pages 163-172
An Integrated Runtime Scheduler for MPI....Pages 173-182
High Performance Checksum Computation for Fault-Tolerant MPI over Infiniband....Pages 183-192
An Evaluation of User-Level Failure Mitigation Support in MPI....Pages 193-203
Efficient MPI Implementation of a Parallel, Stable Merge Algorithm....Pages 204-213
Efficient Distributed Computation of Maximal Exact Matches....Pages 214-223
Scalable Algorithms for Constructing Balanced Spanning Trees on System-Ranked Process Groups....Pages 224-234
A Hybrid Parallelization of Air Quality Model with MPI and OpenMP....Pages 235-245
2 nd Special Session on Improving MPI User and Developer Interaction (IMUDI 2012)....Pages 246-247
A Wish List for Efficient Adjoints of One-Sided MPI Communication....Pages 248-257
On the Usability of the MPI Shared File Pointer Routines....Pages 258-267
Extending MPI to Better Support Multi-application Interaction....Pages 268-274
Versatile Communication Algorithms for Data Analysis....Pages 275-284
High Performance Concurrent Multi-Path Communication for MPI....Pages 285-286
Improving Collectives by User Buffer Relocation....Pages 287-288
Asynchronous Checkpointing by Dedicated Checkpoint Threads....Pages 289-290
Verification of MPI Programs Using Session Types....Pages 291-293
Runtime Support for Adaptive Resource Provisioning in MPI Applications....Pages 294-295
Revisiting Persistent Communication in MPI....Pages 296-297
StarPU-MPI: Task Programming over Clusters of Machines Enhanced with Accelerators....Pages 298-299
Back Matter....Pages -