Program

Registration

Welcome Reception

Registration

Welcome Message

Location: Master Room

Keynote (Joint Keynote with OpenMPCon)

Title: Towards Throughput Oriented Hybrid Programming

Location: Master Room

Session Chair: Antonio J. Peña, Barcelona Supercomputing Center

Technical Program Session 1 - Collective Communication and Topologies

Session Chair: Rolf Rabenseifner, HLRS, University of Stuttgart

Location: Master Room

• Using Node Information to Implement MPI Cartesian Topologies (Best Paper Award Winner)
  William D. Gropp
• Full-Duplex Inter-Group All-to-All Broadcast Algorithms with Optimal Bandwidth
  Qiao Kang, Jesper Larsson Träff, Reda Al-Bahrani, Ankit Agrawal, Alok Choudhary and Wei-Keng Liao
• Optimized Broadcast for Deep Learning Workloads on Dense-GPU InfiniBand Clusters: MPI or NCCL?
  Ammar Ahmad Awan, Ching-Hsiang Chu, Hari Subramoni and Dhabaleswar K. Panda
• Cellular Automata Beyond 100k Cores: MPI vs Fortran Coarrays
  Anton Shterenlikht and Luis Cebamanos

Panel: MPI and Heterogeneity

Session Chair: Jesper Larsson Träff, Vienna University of Technology (TU Wien)

Location: Master Room

Panelists: William Gropp, Martin Schulz, Guillaume Mercier

Technical Program Session 2 - Hybrid Programming

Session Chair: William Gropp, University of Illinois at Urbana-Champaign (UIUC)

Location: Master Room

• Multi-Threading and Lock-Free MPI RMA Based Graph Processing on KNL and POWER Architectures
  Mingzhe Li, Xiaoyi Lu, Hari Subramoni and Dhabaleswar K. Panda
• MPI+OpenMP Tasking Scalability for the Simulation of the Human Brain
  Pedro Valero-Lara, Raül Sirvent, Antonio J. Peña, Xavier Martorell and Jesús Labarta
• Improving the Interoperability between MPI and Task-Based Programming Models
  Kevin Sala, Jorge Bellón, Pau Farré, Xavier Teruel, Josep M. Perez, Antonio J. Peña, Daniel Holmes, Vicenç Beltran, and Jesus Labarta

Social Event - Dinner

The bus to the restaurant:

• Time: 19:00 (Bus will wait till 19:30 the latest)
• Location: In front of C/ Jordi Girona 29. Nexus II building, Barcelona

From the restaurant back to Nexus II building:

• Time: 23:30
• Location: Restaurant Xalet de Montjuïc, Avda. Miramar 31, Barcelona

Registration

EuroMPI 2019

Location: Agora Room

Keynote (Joint Keynote with OpenMPCon)

Title: Sierra, CORAL and OpenMP

Location: Agora Room

Session Chair: Eduard Ayguadé, Barcelona Supercomputing Center

Technical Program Session 3 - Tools and Performance Modeling

Location: Multimedia Room

Session Chair: Hari Subramoni, The Ohio State University

• Improving Performance Models for Irregular Point-to-Point Communication
  Amanda Bienz, William D. Gropp and Luke N. Olson
• Enabling Callback-Driven Runtime Introspection via MPI_T
  Marc-André Hermanns, Nathan T. Hjlem, Michael Knobloch, Kathryn Mohror and Martin Schulz
• MC-CChecker: A Clock-Based Approach to Detect Memory Consistency Errors in MPI One-Sided Applications
  Thanh-Dang Diep, Karl Fürlinger and Nam Thoai
• Performance Model for Mesh Optimization on Distributed-memory Computers
  Domingo Benitez, Jose Escobar, Rafael Montenegro and Eduardo Rodriguez

Poster Session

Location: Multimedia Room

Session Chair: Atsushi Hori, RIKEN

Poster Flash Talks [02:30 pm - 03:00 pm]

• Recent Experiences in Using MPI-3 RMA in the DASH PGAS Runtime
  Joseph Schuchart and José Gracia.
• Improving Support for MPI+OpenMP Applications
  Geoffroy Vallee and Daid E. Bernholdt. [Abstract] [Poster]
• Multithreaded MPI and message matching performance
  Whit Schonbein, Matthew G. F. Dosanjh, Ryan E. Grant and Patrick G. Bridges.
• Toward Developing a Unimem OFI Provider for MPI Support
  Kyunghun Kim, Antonio J. Pena, Paul Carpenter, Polydoros Petrakis, Manolis Ploumidis, Manolis Marazakis, Yanfei Guo, Kenneth Raffenetti and Pavan Balaji. [Abstract] [Poster]
• Making the Case for Portable MPI Process Pinning
  Balazs Gerofi, Rolf Riesen and Yutaka Ishikawa. [Abstract] [Poster]
• A Scalable PMIx Database
  Artem Polyakov, Josh Ladd, Elena Shipunova and Boris Karasev. [Abstract] [Poster]
• MPI Malleability Integration into a Bioinformatics Tool
  Sergio Iserte, Héctor Martínez, Sergio Barrachina, Maribel Castillo, Rafael Mayo, Enrique S. Quintana-Ortí and Antonio J. Peña.
• EPiGRAM-HS: Exascale Programming Models for Heterogenous Systems
  Stefano Markidis and Erwin Laure.
• Topology aware Cartesian grid mapping with MPI
  Christoph Niethammer and Rolf Rabenseifner. [Abstract] [Poster]

Technical Program Session 4 - Fault Tolerance

Location: Multimedia Room

Session Chair: Marc-Andre Hermanns, Forschungszentrum Jülich

• Energy-Efficient Localised Rollback via Data Flow Analysis and Frequency Scaling
  Kiril Dichev, Kirk Cameron and Dimitrios S. Nikolopoulos
• Transparent High-Speed Network Checkpoint/Restart in MPI
  Julien Adam, Jean-Baptiste Besnard, Allen D. Malony, Sameer Shende, Marc Pérache, Patrick Carribault and Julien Jaeger
• MPI Stages: Checkpointing MPI State for Bulk Synchronous Applications
  Nawrin Sultana, Anthony Skjellum, Ignacio Laguna, Matthew Shane Farmer, Kathryn Mohror and Murali Emani

Registration

Industry Session

Location: Multimedia Room

Keynote

Title: A Case for More Adaptivity in MPI

Location: Multimedia Room

Session Chair: Marc Casas, Barcelona Supercomputing Center

Technical Program Session 5 - Performance Optimization

Location: Multimedia Room

Session Chair: Jesper Larsson Träff, Vienna University of Technology

• Efficient Asynchronous Communication Progress for MPI without Dedicated Resources
  Amit Ruhela, Hari Subramoni, Sourav Chakraborty, Mohammadreza Bayatpour, Pouya Kousha and Dhabaleswar K. Panda
• MPI Derived Datatypes: Performance and Portability Issues
  Qingqing Xiong, Purushotham V. Bangalore, Anthony Skjellum and Martin Herbordt
• Using Simulation to Examine the Effect of MPI Message Matching Costs on Application Performance
  Scott Levy and Kurt B. Ferreira
• Supporting MPI-Distributed Stream Parallel Patterns in GrPPI
  Javier Fernández Muñoz, Manuel F. Dolz, David del Rio Astorga, Javier Prieto Cepeda and J. Daniel García

MareNostrum Visit