EuroMPI 2011 Program

Download as PDF


EuroMPI 2011 at a glance
Monday 19th September
Tuesday 20th September
Wednesday 21st September

Monday 19th Tuesday 20th Wednesday 21st
8:30-9:20 Registration
9:20-9:30 Opening Session
9:30-10:15 Invited Talk
Pete Beckman
10:15-10:45 Coffee
9:00-9:45 Invited Talk
Michael Resch
9:45-10:15 Best Paper
10:15-10:45 Coffee
9:30-10:15 Invited Talk
Torsten Hoefler
10:15-10:45 Coffee
10:45-12:30 1
12:30-1:30 Lunch
01:30-02:15 Invited Talk
George Bosilca
2:15-3:30 2
3:30-4:00 Coffee
4:00-5:15 3
8:00- +Dinner at Dionysos Restaurant
4 5
12:30-1:30 Lunch
1:30-2:15 Invited Talk
Sudip Dosanjh
2:15-2:45 Best Paper
2:45-3:35 6
5:30- *Excursion to Oia and Dinner
12:30-1:30 Lunch
1:30-2:15 Invited Talk
Michael Kagan
8 Posters
1Communication Issues and Improvements
2I/O Issues and Improvements
3Networking Issues and Improvements
4Algorithms and Tools
5Interaction with Hardware
6Implementation Issues and Improvements
7Applications and Performance Evaluation
8Fault Tolerance

+Price for accompanied persons: 42 euros per person.
*Price for accompanied persons: 78 euros per person.

Monday 19th September
8:30-9:20 Registration
9:20-9:30 Opening Session
9:30-10:15 Invited Talk:
Pete Beckman, Director, Exascale Technology and Computing Institute, "Exascale System Software Challenges – From Bare Metal to Applications"
10:15-10:45 Coffee
10:45-12:30 Session 1: Communication Issues and Improvements
Enes Bajrovic and Jesper Larsson Träff, “Using MPI derived datatypes in numerical libraries”
Guillaume Mercier and Emmanuel Jeannot, “Improving MPI Applications Performance on Multicore Clusters with Rank Reordering”
Mohammad Javad Rashti, Jonathan Green, Pavan Balaji, Ahmad Afsahi and William Gropp, “Multi-core and Network Aware MPI Topology Functions”
Carl Albing, Norm Troullier, Stephen Whalen, Ryan Olson, Howard Pritchard, Joe Glenski and Hugo Mills , “Scalable Node Allocation for Improved Performance in Regular and Anisotropic 3D Torus Supercomputers”
12:30-1:30 Lunch
1:30-2:15 Invited Talk:
George Bosilca, Research Professor, Innovative Computing Laboratory, UTK, "Will MPI Remain Relevant?"
2:15-3:30 Session 2: I/O Issues and Improvements
Chen Jin, Saba Sehrish, Wei-Keng Liao, Alok Choudhary and Karen Schuchardt, “Improving the Average Response Time in Collective I/O”
Mohamad Chaarawi, Edgar Gabriel, Rainer Keller, Richard Graham, George Bosilca and Jack Dongarra, “OMPIO: A Modular Software Architecture for MPI I/O”
Vishwanath Venkatesan, Mohamad Chaarawi, Edgar Gabriel and Torsten Hoefler, “Design and Evaluation of Nonblocking Collective I/O Operations”
3:30-4:00 Coffee
4:00-5:15 Session 3: Networking Issues and Improvements
Sreeram Potluri, Hao Wang, Vijay Dhanraj, Sayantan Sur and Dhabaleswar Panda, “Optimizing MPI One Sided Communication on Multi-core InfiniBand Clusters using Shared Memory Backed Windows”
Howard Pritchard, Igor Gorodetsky and Darius Buntinas, “A uGNI-Based MPICH2 Nemesis Network Module for Cray XE Computer Systems”
Brian Barrett, Ron Brightwell, Scott Hemmert, Kyle Wheeler and Keith Underwood, “Using Triggered Operations to Offload Rendezvous Messages”
8:00- Dinner at Dionysos Restaurant (price for accompanied persons: 42 euros per person)
Tuesday 20th September
9:00-9:45 Invited Talk:
Michael Resch, Director, High Performance Computing Center Stuttgart (HLRS), "Experience of a PRACE Center"
9:45-10:15 Best Paper Session
Tobias Hilbrich, Matthias S. Mueller, Martin Schulz and Bronis R. de Supinski, “Order Preserving Event Aggregation in TBONs”
10:15-10:45 Coffee
10:45-12:30 Session 4: Algorithms and Tools Session 5: Interaction with Hardware
William D Gropp, Torsten Hoefler, Rajeev Thakur and Jesper Larsson Träff, “Performance Expectations and Guidelines for MPI Derived Datatypes” Vaibhav Sundriyal and Masha Sosonkina, “Per-call Energy Saving Strategies in All-to-all Communications”
Alexey Salnikov, Dmitry Andreev and Roman Lebedev, “The analysis of cluster interconnect with the network_tests2 toolkit” Jerome Soumagne, John Biddiscombe and Aurélien Esnard, “Data Redistribution using One-sided Transfers to In-memory HDF5 Files”
Christian Siebert and Felix Wolf, “Parallel Sorting with Minimal Data” Isaías A. Comprés Ureña, Michael Riepen and Michael Konow, “RCKMPI – Lightweight MPI Implementation for Intel’s Single-Chip Cloud Computer (SCC)”
Markus Geimer, Marc-Andre Hermanns, Christian Siebert, Felix Wolf and Brian Wylie, “Scaling performance tool MPI communicator management”
12:30-1:30 Lunch
1:30-2:15 Invited Talk:
Sudip Dosanjh, Senior Manager, Sandia National Laboratories, "Achieving Exascale Computing Through Hardware/Software Co-design"
2:15-2:45 Best Paper Session
Adam Moody, Dong Ahn and Bronis R. de Supinski, “Exascale Algorithms for Generalized MPI_Comm_split”
2:45-3:35 Session 6: Implementation Issues and Improvements
Rune Bromer, Frederik Hantho and Brian Vinter, “pupyMPI - MPI Implemented in Pure Python”
David Goodell, William Gropp, Xin Zhao and Rajeev Thakur, “Scalable Memory Use in MPI: A Case Study with MPICH2”
5:30- Excursion to Oia and Dinner (price for accompanied persons: 78 euros per person)
Wednesday 21st September
9:30-10:15 Invited Talk:
Torsten Hoefler, University of Illinois, Urbana-Champaign, "Writing Parallel Libraries with MPI - Common Practice, Issues, and Extensions"
10:15-10:45 Coffee
10:45-12:30 Session 7: Applications and Performance Evaluation IMUDI Workshop
Juan Sillero, Guillem Borrell I Nogueras, Javier Jimenez and Robert D. Moser, “Hybrid OpenMP-MPI Turbulent Boundary Layer Code over 32k cores” James Dinan, Sriram Krishnamoorthy, Pavan Balaji, Jeff R. Hammond, Manojkumar Krishnan, Vinod Tipparaju, and Abhinav Vishnu, “Noncollective Communicator Creation in MPI”
Manuel Hasert, Harald Klimach and Sabine Roller, “CAF versus MPI - Applicability of Coarray Fortran to a Flow Solver” Matti Bickel, Adrian Knoth and Mladen Berekovic, “Evaluation of Interpreted Languages with Open MPI”
Kevin T. Pedretti, Ron Brightwell, Doug Doerfler, K. Scott Hemmert and James H. Laros III, “The Impact of Injection Bandwidth Performance on Application Scalability” Simone Pellegrini, Radu Prodan, and Thomas Fahringer, “Leveraging C++ Meta-Programming Capabilities to Simplify the Message Passing Programming Model”
Teng Ma, Aurelien Bouteiller, George Bosilca and Jack Dongarra, “Impact of Kernel-Assisted MPI Communication over Scientific Applications: CPMD and FFTW”
12:30-1:30 Lunch
1:30-2:15 Invited Talk:
Michael Kagan, Melanox, "Paving the Road to Exascale Computing"
2:15-4:00 Session 8: Fault Tolerance Posters: Short presentations
Joshua Hursey, Thomas Naughton, Geoffroy Vallee and Richard Graham, “A Log-Scaling Fault Tolerant Agreement Algorithm for a Fault Tolerant MPI” Jason Cope, Kamil Iskra, Dries Kimpe and Robert Ross, “Portable and Scalable MPI Shared File Pointers“
Alexandre Gonçalves, Matheus Bersot, André Bulcão, Cristina Boeres, Lúcia Drummond and Vinod Rebello, “Fault Tolerance in an Industrial Seismic Processing Application for Multicore Clusters” Sreeram Potluri, Sayantan Sur, Devendar Bureddy and Dhabaleswar K. Panda, “Design and Implementation of Key Proposed MPI-3 One-Sided Communication Semantics on InfiniBand”
Kurt Ferreira, Rolf Riesen, Ron Brightwell, Patrick Bridges and Dorian Arnold, “Libhashckpt: Hash-based Incremental Checkpointing Using GPU's” Carsten Clauss, Stefan Lankes and Thomas Bemmerl, “Performance Tuning of SCC-MPICH by means of the Proposed MPI-3.0 Tool Interface”
Posters: Short presentations continue below
Kiril Dichev, Alexey Lastovetsky and Vladimir Rychkov,“Improvement of the Bandwidth of Cross-Site MPI Communication Using Optical Fiber”
Darius Buntinas, “Scalable Distributed Consensus to Support MPI Fault Tolerance”
Joshua Hursey, Richard Graham, Greg Bronevetsky, Darius Buntinas, Howard Pritchard & David Solt, “Run-Through Stabilization: An MPI Proposal for Process Fault Tolerance”
Yonghong Yan, Sanjay Chatterjee, Zoran Budimli and Vivek Sarkar, “Integrating MPI with Asynchronous Task Parallelism using Habanero-C”
Juan-Antonio Rico-Gallego and Juan-Carlos Díaz-Martín, “Performance Evaluation of Thread-Based MPI in Shared Memory”
Edward Givelberg, Alexander Szalay, Kalin Kanov & Randal Burns, “MPI-DB, A Parallel Database Services Software Library for Scientific Computing”
George Bosilca, Thomas Herault, Pierre Lemarinier, Ala Rezmerita and Jack J. Dongarra, “Scalable Runtime for MPI: Efficiently Building the Communication Infrastructure”