Tehdyt toimenpiteet
Program
Monday Sept. 7
08.30 Registration opens
09:00-10:30 Tutorial starts: Practical Formal Verification of MPI and Thread Programs
Ganesh Gopalakrishnan, Robert M. Kirby, and Anh Vo
- Overview of Formal Verification (Ganesh Gopalakrishnan)
- Overview of MPI and some MPI constructs (Mike Kirby)
- Demo of ISP, our Dynamic Verification tool for MPI (Ganesh)
- Architecture of ISP
- Present Any_src_can_deadlock with hardwired timings under MPIRUN and under ISP
- In-depth presentation of POE-Illustration
- Presentation of traces by Program Order, Internal Issue Order, and Time Order
- Help Attendees Boot LiveCD/USB and run POE Illustration (Anh Vo)
11:00-12:30
- Advanced MPI Verification (Ganesh)
- Illustration of more deadlocks
- Resource Leak Detection
- How ISP reveals the details of Iprobe (receive need not match the probed)
- Assertion Violation Detection
- Theoretical Concepts behind ISP’s POE Algorithm – how the algorithm directly follows our semantics
12:00-12:30 Matrix Multiplication (Mike)
- Four versions of Matrix Multiplication
- Audience participation in running Dynamic Verification
- Audience participation in reviewing results using ISP user-interfaces
14:00-15:30 Tool Assisted Verification using ISP: We assign problems and help you solve them (all tutors)
Dynamic Verification of Shared Memory Thread Programs using Inspect (Gopalakrishnan)
15:30-16:00 Coffee Break
16:00-16:30 Tool Assisted Verification using Inspect (all tutors)
16:30-17:00 Conclusions, Future Work
Tuesday Sept. 8
08:30-09:15 Registration opens + coffee
09:15-09:30 Welcome,
09:30-10:15 Invited talk "Real-Time Message Compression in Software",
Martin Burtscher, University of Texas,USA
10:15-10:30 Coffee break
10:30-11:30 Parallel sessions at Luna Auditorium and Castor by the Conference Lobby, Spektri Duo
Sesson1.1: Performance
Rosa Filgueira, David Exposito, Alejandro Calderón and Jesus Carretero. CoMPI:
Enhancing MPI based applications performance and scalability using run-time compression
Stephen F. Siegel and Andrew R. Siegel: A Memory-Efficient Data Redistribution Algorithm
Session1.2: Fault Tolerance
Leonardo Fialho, Guna Santos, Angelo Duarte, Dolores Rexachs and Emilio Luque:
Challenges and Issues of the Integration of RADIC into Open MPI
Gang Wang, Xiaoguang Liu, Ang Li and Fan Zhang:
In-Memory Checkpointing for MPI Programs by XOR-based Double-Erasure Codes
11:30-13:00 Lunch break
13:00-14:00 Parallel sessions
Session2.1: MPI I/O
Faisal Ghias Mir and Jesper Larsson Träff:
Exploiting efficient Transpacking for One-sided Communication and MPI-IO
Javier García Blas, Florin Isaila, Jesús Carretero, Robert Latham and Robert Ross:
Multiple-level MPI file write-back and prefetching for Blue Gene systems
Session2.2: Performance
Simone Pellegrini, Jie Wang, Thomas Fahringer and Hans Moritsch:
Optimizing MPI Runtime Parameter Settings by Using Machine Learning
Jaidev Sridhar and Dhabaleswar Panda:
Impact of Node Level Caching in MPI Job Launch Mechanisms
14:00-14:45 Invited talk "The MPI 2.2 standard and the emerging MPI 3 standard",
Richard Graham, Oak Ridge National Laboratory, USA
14:45-15:00 Coffee break
15:00-16:00 Parallel sessions
Session3.1: MPI I/O
David Buettner, Julian Kunkel and Thomas Ludwig:
Using Non-Blocking I/O Operations in High Performance Computing to Reduce Execution Times
Saba Sehrish, Jun Wang and Rajeev Thakur:
Conflict Detection Algorithm to Minimize Locking for MPI-IO Atomicity
Session3.2: OPENMP
Damián A. Mallón, Guillermo Taboada, Carlos Teijeiro Barjas, Juan Touriño, Basilio B. Fraguela,
Andrés Gómez, Ramón Doallo and J. Carlos Mouriño:
Performance Evaluation of MPI, UPC and OpenMP on Multicore Architectures
Francisco de Sande, Ruymán Reyes, Antonio J. Dorta and Francisco Almeida:
Automatic Hybrid MPI+OpenMP code generation with llc
16:30-17:00 Posters
Justin Wozniak, Bryan Jacobs, Robert Latham, Sam Lang, Seung Woo Son and Robert Ross:
Implementing Reliable Data Structures for MPI Services in High Component Count Systems
Mónica M. Denham, Ana Cortés and Tomás Margalef:
Parallel Dynamic Data Driven Genetic Algorithm for Forest Fire Prediction
Hao Zhu, David Goodell, William Gropp and Rajeev Thakur:
Hierarchical Collectives in MPICH2
Juan-Carlos Díaz-Martín, Juan-Antonio Rico-Gallego, Jesús-María Álvarez-Llorente and
Francisco-José Perogil-Duque:
An MPI-1 Compliant Thread-based Implementation
Sarvani Vakkalanka, Grzegorz Szubzda, Anh Vo, Ganesh Gopalakrishnan, Robert M. Kirby
and Rajeev Thakur:
Static-analysis Assisted Dynamic Verification to Efficiently Handle Waitany Non-determinism
17:00-20:00 Welcoming Party & Dinner
Wednesday Sept. 9
08:30 Registration opens
09:00-09:15 Organizational remarks
09:15-10:00 Invited talk "MPI at Exascale: Challenges for Data Structures and Algorithms",
William Gropp, University of Illinois Urbana-Champaign, USA
10:00-10:30 Coffee break
10:30-11:30 Parallel sessions
Session4.1: Programming Paradigms and Collective Operations
Torsten Hoefler, Andrew Lumsdaine and Jack Dongarra:
Towards Efficient MapReduce Using MPI
Ying Qian and Ahmad Afsahi:
Process Arrival Pattern Aware Alltoall on InfiniBand
Session4.2: Applications
Michael Hofmann and Gudula Ruenger:
Fine-grained Data Distribution Operations for Particle Codes
Lars Schneidenbach, Bettina Schnor, Martin Gebser, Roland Kaminski, Benjamin Kaufmann
and Torsten Schaub:
Experiences running a Parallel Answer Set Solver on Blue Gene
11:30-13:00 Lunch break
13:00-14:00 Parallel sessions
Session5.1: Library internals
Richard Graham and Rainer Keller: Dynamic Communicators in MPI
Troy LeBlanc, Rakhi Anand, Edgar Gabriel and Jaspal Subhlok:
VolpexMPI: an MPI Library for Execution of Parallel Applications on Volatile Nodes
Session5.2: MPI on Multicore Architectures
Marc Perache, Patrick Carribault and Herve Jourdren:
MPC-MPI: An MPI Implementation Reducing the Overall Memory Consumption
Guillaume Mercier and Jérôme Clet-Ortega:
Towards an Efficient Process Placement Policy for MPI Applications in Multicore Environments
14:00-16:00 Parallel sessions
Session6.1: ParSim
Michael Bader, Miriam Mehl, Tobias Weizierl:
Towards Parallel Simulation Environments using Recursively Structured Adaptive Grids and Space-filling Curves
C.H. Moreno-Montiel, F. Rojas-González, G. Román-Alonso, S. Cordero-Sánchez, M.A. Castro-García and M. Aguilar-Cornejo:
A Parallel Simulator for Mercury (Hg) Porosimetry
Ralf-Peter Mundani, Alexander Düster, Jovana Knezevic, Andreas Niggl, and Ernst Rank:
Dynamic Load Balancing Strategies for Hierarchical p-FEM Solvers
Peng Zeng, Samuel Sarholz, Christian Iwainsky, Bernd Binninger, Norbert Peters, and Marcus Herrmann:
Simulation of Primary Breakup for Diesel Spray with Phase Transition
14:45-15:00 Coffee break
Session6.2: Verification of MPI Programs
Sriram Aananthakrishnan, Michael DeLisi, Sarvani Vakkalanka, Anh Vo, Ganesh Gopalakrishnan,
Robert Kirby and Rajeev Thakur:
How Formal Dynamic Verification Tools Facilitate Novel Concurrency Visualizations
Anh Vo, Sarvani Vakkalanka, Jason Williams, Ganesh Gopalakrishnan, Robert Kirby and
Rajeev Thakur:
Sound and Efficient Dynamic Verification of MPI Programs with Probe Non-Determinis
Chris Gottbrath: Bringing Reverse Debugging to HPC
16:00-17:00 Invited talk "Model-based Optimization of MPI Collective Operations for Computational Clusters"
Alexey Lastovetsky, University College Dublin, Ireland
17:10 Bus transportation to the Hanasaari pier
17:30-20:00 Boat trip & Dinner on board
Thursday Sept. 10
08:30 Registration opens + coffee
09:15-10:00 Invited talk,"Using MPI to Implement Scalable Libraries"
Ewing Lusk, Argonne National Laboratory, USA
10:00-10:30 Coffee break
10:30-11:30 Outstanding papers 1
Pavan Balaji, Darius Buntinas, David Goodell, William Gropp, Sameer Kumar, Ewing Lusk,
Rajeev Thakur and Jesper Larsson Traeff:
MPI on a Million Processors
Marc-Andre Hermanns, Markus Geimer, Bernd Mohr and Felix Wolf:
Scalable Detection of MPI-2 Remote Memory Access Inefficiency Patterns
11:30-13:00 Lunch break
13:00-14:00 Outstanding papers 2
Robert Ross, Robert Latham, William Gropp, Ewing Lusk and Rajeev Thakur:
Processing MPI Datatypes Outside MPI
Shinji Sumimoto, Kohta Nakashima, Akira Naruse, Kouichi Kumon, Takashi Yasui,
Yoshikazu Kamoshida, Hiroya Matsuba, Atsushi Hori and Yutaka Ishikawa:
The Design of Seamless MPI Computing Environment for Commodity-based Clusters
14:00-15:00 Invited talk, "Formal Verification for Scientific Computing: Trends and
Progress",
Stephen Siegel, University of Delaware, USA
15:00-1510 Coffee break
15:10-16:00 MPI forum information session
16:00-17:00 Open forum