9th JLESC workshop, Knoxville USA (April 15-17 2019)


9th JLESC Workshop

Objectives 

The workshop gathers leading researchers in high-performance computing from the JLESC partners INRIA, the University of Illinois, Argonne National Laboratory, Barcelona Supercomputing Center, Jülich Supercomputing Centre, RIKEN R-CCS and The University of Tennessee to explore the most recent and critical issues in advancing the field of HPC from petascale to the extreme scale era.

The workshop will feature sessions on these seven central topics:

  • Applications and mini-apps
  • Parallel Programming models and runtime
  • Performance tools
  • Resilience
  • Big Data, I/O and in-situ visualization
  • Numerical methods and algorithms
  • Advanced architectures

In addition, dedicated sessions on computational fluid dynamics, computational biology and climate/weather research are planned.

A key objective of the workshop is to identify new research collaborations and establish a roadmap for their implementation.

The workshop is open to Illinois, INRIA, ANL, BSC, JSC, Riken R-CCS and UTK faculties, researchers, engineers and students who want to learn more about Post-Petascale / Pre-Exascale Computing.

4th floor UT Conference Center

UT Conference Center 4th floort

Agenda 


Track 1 Track 2 Track 3
07:30 Registration & Breakfast
08:30 Opening
Robert Speck, Yves Robert, Jack Dongarra and George Bosilca, UTK
Franck Cappello, ANL Marc Snir, UIUC
09:00 Plenary: Delivering on the Exascale Computing Project Mission for the U.S. Department of Energy
Kothe, Douglas B (ORNL)
Session chair: Jack Dongarra
09:30
10:00 Break
10:30
Session chair: Atsushi Hori

Session chair: Jakub Kurzak

Session chair: Michela Taufer
11:00
11:30
12:00 Lunch
12:30
13:00
13:30
session moved on Wednesday (PTM3.1)

Session chair: Robert Speck
Session organizer: Piotr Luszczek
14:00
14:30
15:00 Break
15:30
Session chair: Daniel Katz

Session chair: Gabriel Antoniu

Session chair: Stan Tomov
16:00
16:30
17:00
17:30 Adjourn
18:00
18:30
19:00 Social Event Dinner at Calhoun's on the River (upstairs)

Track 1 Track 2 Track 3
07:30 Breakfast
08:30 Plenary: Not Your Grandfather’s Tractor – How AI & IoT Are Transforming Production Agriculture
Mark Moran (John Deere)
Session chair: Brendan McGinty
09:00
09:30 Break
10:00
Session chair: Hartwig Anzt

Session chair: Rosa Badia
ARM (more info)
Session organizer: Mitsuhisa Sato
10:30
11:00
11:30
12:00 Lunch
12:30
13:00
13:30
Session chair: Christian Perez
ARM (more info)
Session organizer: Mitsuhisa Sato
14:00 Session organizer: Kazutomo (Kaz) Yoshii
14:30
15:00
15:30
16:00 Poster Sessions @ Hilton Hotel, Hiwassee Room (Lobby level)
16:30
17:00
17:30 Social event: Riverboat Dinner and Celebration (meeting at the dock at 17:50, the boat leaves at 18:00 SHARP)
18:00
18:30
19:00

Track 1 Track 2 Track 3
07:30 Breakfast
08:30 Plenary: Programming workflows for Advanced Cyberinfrastructure Platforms
Rosa Badia (BSC)
Session chair: Franck Cappello
09:00
09:30 Break
10:00
Session chair: Ruth Schoebel

Session chair: Bogdan Nicolae
Session organizer: Laxmikant (Sanjay) Kale
10:30
11:00
11:30
12:00 Lunch
12:30
13:00
13:30
Session chair: Jon Calhouns

Session chair: Ana Gainaru
14:00
14:30
15:00 Closing
15:30
16:00
16:30
17:00
17:30
19:00 Closing Dinner: Lonesome Dove

Agenda Items 


Title Presenter
Not Your Grandfather’s Tractor – How AI & IoT Are Transforming Production AgricultureMark Moran
John Deere
Abstract:

After some background on the sweeping changes happening in agricultural technology as a result of General Purpose Technologies, Mark will discuss how AI and IoT are merging on the farm, creating some compelling problems to be solved on the edge.

Programming workflows for Advanced Cyberinfrastructure PlatformsRosa M. Badia
Barcelona Supercomputing Center
Abstract:

In the design of an advanced cyberinfrastructure platform (ACP), that involve sensors, edge devices, instruments, computing power in the cloud and HPC, a key aspect is how to describe the applications to be executed in such platform.

Very often these applications are not standalone, but involve a set of sub-applications or steps composing a workflow. The scientists then rely on effective environments to describe their workflows and engines to manage them in complex infrastructures.

COMPSs is a task-based programming model that enables the development of workflows that can be executed in parallel in distributed computing platform. The workflows that we currently support may involve different types of tasks, such as parallel simulations (MPI) or analytics (i.e., written in Python thanks to PyCOMPSs, the Python binding for COMPSs). COMPSs, through and storage interface, makes transparent the access to persistent data stored in key-value databases (Hecuba) or object-oriented distributed storage environments (dataClay).

While COMPSs has been developed from its early times for highly distributed environments, we have been extending it to deal with more challenging environments, with edge devices and components in the fog, that can appear and disappear.

Delivering on the Exascale Computing Project Mission for the U.S. Department of EnergyDouglas B. (Doug) Kothe
Oak ridge National Laboratory
Abstract:

The vision of the U.S. Department of Energy (DOE) Exascale Computing Project (ECP), initiated in 2016 as a formal DOE project executing through 2023, is to accelerate innovation with exascale simulation and data science solutions that enhance U.S. economic competitiveness, strengthen our national security, and change our quality of life. ECP’s mission is to deliver exascale-ready applications and solutions that address currently intractable problems of strategic importance and national interest; create and deploy an expanded and vertically integrated software stack on DOE HPC exascale and pre-exascale systems, thereby defining the enduring US exascale ecosystem; and leverage U.S. HPC vendor research activities and products into DOE HPC exascale systems. The project is a joint effort of two DOE programs: the Office of Science Advanced Scientific Computing Research Program and the National Nuclear Security Administration Advanced Simulation and Computing Program. ECP’s RD&D activities are carried out by over 100 teams of scientists and engineers from the DOE national laboratories, universities, and U.S. industry. These teams have been working together since the fall of 2016 on the development of applications, software technologies, and hardware technologies and architectures:

Applications: Creating or enhancing the predictive capability of applications through algorithmic and software advances via co-design centers; targeted development of requirements-based models, algorithms, and methods; systematic improvement of exascale system readiness and utilization; and demonstration and assessment of effective software integration.

Software Technologies: Developing and delivering a vertically integrated software stack containing advanced mathematical libraries, extreme-scale programming environments, development tools, visualization libraries, and the software infrastructure to support large-scale data management and data science for science and security applications.

Hardware and Integration: Supporting U.S. HPC vendor R&D focused on innovative architectures for competitive exascale system designs; objectively evaluating hardware designs; deploying an integrated and continuously tested exascale software ecosystem at DOE HPC facilities; accelerating application readiness on targeted exascale architectures; and training on key ECP technologies to accelerate the software development cycle and optimize productivity of application and software developers.

Illustrative examples will be given on how the ECP teams are delivering in these three areas of technical focus, with specific emphasis on recent work on the world’s #1 supercomputer, namely the Summit leadership system at Oak Ridge National Laboratory.

Title Presenter Project Abstract
PT M1.1Comparing the performance of rigid, moldable and grid-shaped applications on failure-prone HPC platformsThomas HeraultOptimization of Fault-Tolerance Strategies for Workflow Applications
Monday 10:30
Using Lossy Compression for Linear Solver ResilienceMirco AltenberndEffective Use of Lossy Compression for Numerical Linear Algebra Resilience and Performance
Exploring the Variability and Error Distributions of Lossy Compression AlgorithmsJon Calhoun (Clemson University)Checkpoint/Restart of/from lossy state
Reducing the memory footprint in Krylov methods through lossy compressionNick Schenkels (Inria) Checkpoint/Restart of/from lossy state
PT M1.2Review on standard eigensolvers on a high-end GPU systemToshiyuki Imamura (RIKEN)HPC libraries for solving dense symmetric eigenvalue problems
Monday 10:30
Adding Inter-Node Communication to a C++ Tasking Framework for Synchronization-Critical ApplicationsIvo Kabadshow (JSC)Scalability Enhancements to FMM for Molecular Dynamics Simulations
Fast Integrators for Scalable Quantum Molecular DynamicsAlina Kononov (UIUC)Fast Integrators for Scalable Quantum Molecular Dynamics
Solving the Bethe-Salpeter eigenvalue problem on thousands of coresXiao Zhang (University of Illinois)The ChASE library for large Hermitian eigenvalue problems
PT M1.3Evaluating OmpSs2 Runtime for XMP 2.0 PGAS Task ModelJinpil Lee (RIKEN R-CCS)Sharing and extension of OmpSs2 runtime for XMP 2.0 PGAS task model
Monday 10:30
Analyses and Transformations for Reverse Mode Differentiation of OpenMPJan Hueckelheim (Imperial College London / Argonne)Shared Infrastructure for Source Transformation Automatic Differentiation
Performance of S-Step and Pipelined Krylov MethodsIchitaro Yamazaki (UTK)Reducing Communication in Sparse Iterative and Direct Solvers
Designing Scalable Solvers for Enlarged Krylov Subspace MethodsShelby Lockhart (University of Illinois)Reducing Communication in Sparse Iterative and Direct Solvers
PT A1.2Project update: Developer tools for porting and tuning parallel applications on extreme-scale parallel systemsBrian Wylie (JSC)Developer tools for porting and tuning parallel applications on extreme-scale parallel systems
Monday 13:30
On the Applicability of PEBS based Online Memory Access Tracking for Heterogeneous Memory Management at ScaleLeonardo Bautista (BSC)Deep Memory Hierarchies
Monitoring MPI and forecasting network usageBogdan Nicolae (ANL)Towards accurate network utilization forecasting using portable MPI-level monitoring
Preliminary Result of MPI International SurveyAtsushi Hori (Riken)MPI International Survey
PT A3.1Rescaling Transient Distributed Storage SystemsNathanaël CheriereReconfiguring Distributed Storage Systems on HPC infrastructures
Wednesday 13:30
Experiment Management and Repeatability Services in ChameleonKate KeaheyAdvancing Chameleon and Grid'5000 testbeds
Improving the Performance and Energy Efficiency of HPC Applications Using Autonomic Computing TechniquesEric Rutten, InriaImproving the Performance and Energy Efficiency of HPC Applications Using Autonomic Computing Techniques
Leveraging Blob-Based Storage for Stream ProcessingPierre Matri, ANLTowards Blob-Based Convergence Between HPC and Big Data
PT A3.2Heterogeneous Hierarchical Workflow CompositionOrçun YILDIZ (ANL)Extreme-Scale Workflow Tools - Swift, Decaf, Damaris, FlowVR
Wednesday 13:30
Evaluating OpenCL Kernels for FPGA platformsZheming JinEvaluating high-level programming models for FPGA platforms
Cultivating a community about FPGA-HPC platformsKazutomo Yoshii (ANL)Evaluating high-level programming models for FPGA platforms
Status of the Simplified Sustained System performance benchmark ProjectMiwako Tsuji (RIKEN)Simplified Sustained System performance benchmark
Toward MPI-based Workflow Execution: Applications and StandardizationJustin M. Wozniak (ANL)Resource Management, Scheduling, and Fault-Tolerance for HPC Workflows
Title Presenter Topic
ST A1.1Scheduling stochastic tasks under buget/deadline constraintsYves RobertProgramming Languages and Runtimes
Monday 15:30
Tasking Meets GPUs: Fighting Deadlocks and Other MonstersLaura Morgenstern, JSCNumerical Methods, Programming Languages and Runtimes
Python-based workflows & software sustainabilityDaniel S. Katz (UIUC)Programming Languages and Runtimes
Scheduling stochastic jobs on HPC platformsHongyang Sun (INRIA)Programming Languages and Runtimes
Runtime Systems to Support GPU Acceleration of Irregular ProblemsThomas Herault (UTK)Performance Tools 2 Programming Languages and Runtimes
Scalable Dynamic DAG Building for data-flow task-based runtime systemsReazul Hoque (UTK)Programming Languages and Runtimes
Integrating Futures as Async Provider into PaRSEC Runtime SystemYu Pei (UTK)Programming Languages and Runtimes
ST A1.2Evaluating High-throughput Applications Models for in situ AnalyticsValentin Honoré (Univ. Bordeaux/Inria)I/O, Storage and In-Situ Processing
Monday 15:30
Scalable Data Ingestion for Stream ProcessingGabriel AntoniuI/O, Storage and In-Situ Processing
Combining Checkpointing and Replication for Reliable Execution of Linear WorkflowsValentin Le Fevre (Inria)Resilience
Optimal algorithms for Adjoint ComputationJulien Herrmann (Inria Bordeaux)I/O, Storage and In-Situ Processing
Analytics4MD: In-Situ Data Analytics for Next Generation Molecular Dynamics Workflows.Stephen Thomas (UTK)I/O, Storage and In-Situ Processing, Apps & Mini-Apps
Analytics4MD: In-Situ Data Analytics for Next Generation Molecular Dynamics Workflows.Stephen Thomas (UTK)I/O, Storage and In-Situ Processing, Apps & Mini-Apps
Developing Holistic Measurement Driven System Intelligent AssessmentBill Kramer (NCSA)Architecture, Resilience
ST A1.3Scalable MPI completionThananon Patinyasakdikul (UTK)Programming Languages and Runtimes
Monday 15:30
Improved Convolution Implementations on NVIDIA GPUsMarc Jorda (BSC)Numerical Methods; Programming Languages and Runtimes
Towards Continuous Benchmarking (CB) for sustainable software developmentHartwig Anzt (UTK)PDE solvers for CFD, data-sparse and sparse formats, GPU asynchronous programming, preconditioned iterative methods with optimal communication
Classification methods for Polar Stratospheric CloudsRocco SedonaNumerical Methods
BONSAI: Benchmarking OpeN Software Autotuning InfrastructureJakub Kurzak (UTK)Performance Tools
The HeAT FrameworkDaniel Gage Coquelin (JSC)Advanced Architectures
DeepHyper: Scalable Asynchronous Neural Architecture and Hyperparameter Search for Deep Neural NetworksRomit Maulik (ANL)Machine Learning
ST M2.1HPC Computations for Respiratory Flows - Towards Personalized MedicineAndreas Lintermann (JSC)Numerical Methods
Tuesday 10:00
MagmaDNN - High-Performance Data Analytics for Manycore GPUs and CPUsKwai Wong (UTK)Numerical Methods
SLATE: Software for Linear Algebra Targeting ExascaleJakub Kurzak (UTK)Numerical Methods
Mixed-precision solversAzzam Haidar (NVidia)Numerical Methods
ParILUT - A Parallel Threshold ILU for Manycore ParallelismHartwig Anzt (UTK)PDE solvers for CFD, data-sparse and sparse formats, GPU asynchronous programming, preconditioned iterative methods with optimal communication
Adaptive Precision PreconditioningHartwig Anzt (UTK)PDE solvers for CFD, data-sparse and sparse formats, Adaptive precision Preconditioning
Mesh partitioning and load balancing on heterogeneous systemsRicard Borrell (BSC)Numerical Methods
ST M2.2Characterization of Power and Performance in Data-Intensive Applications using MapReduce over MPIMichela Taufer (UTK)Apps & Mini-Apps; I/O, Storage and In-Situ Processing
Tuesday 10:00
POP2 - The Quest for Performance ContinuesBernd Mohr (JSC)Performance Tools
Understanding native event semanticsAnthony Danalis (UTK)Performance Tools
Using Software-based Performance Counters to Analyze MPI PerformanceDavid Eberius (UTK)Performance tools
A design space exploration targeting next-generation HPC architecturesConstantino Gómez (BSC)Advanced Architecture
Exploring Next Generation High Performance Computing Architecture for Post-Moore EraMasaaki Kondo (RIKEN-CCS)Advanced Architectures, Performance Tools
ST A2.2High-Performance Custom Computing System with Stratix10 FPGAsKentaro Sano, Riken R-CCSAdvanced Architectures
Tuesday 13:30
Dynamic Resource Partitioning for Complex Workloads using ContainersSwann Perarnau (ANL)Advanced Architectures, Performance Tools, Apps & Mini-apps
Improving the performance of a AI-based Big Data application using Edge computingPedro Silva (INSA de Rennes / Inria)Advanced Architectures
The Load Balancing SandboxVincent ReverdyParallel Programming models and runtime, Applications and mini-apps
Downscaling Soil Moisture Data from Remote SensingDanny Rorabaugh (UTK)Apps & Mini-Apps
FFT for large-scale heterogeneous systemsAlan Ayala (UTK)Apps & Mini-Apps
Scalable graph algorithms for cluster-findingLaxmikant Kale and Raghavendra KanakagiriApps & Mini-Apps
Porting Applications towards Modular SupercomputingAlexander Sebastian Achilles (JSC)Advanced Architectures, Apps & Mini-Apps
ST M3.1Supporting multiple virtual ranks in MPI: exploration of techniques including process-in-processLaxmikan Kale and Same White (UIUC)Programming Languages and Runtimes
Wednesday 10:00
Performance and Power Consumption Analysis of Arm Scalable Vector Extension with gem5 SimulatorTetsuya Odajima, (RIKEN R-CCS)Performance Tools
On the Impact of OpenMP Task GranularityChristian Perez (Inria)Programming Languages and Runtimes
SLATE-PaRSEC interface, Bridging Two Dataflow RepresentationsDamien Genet (UTK)Programming Languages and Runtimes
Communication runtime for asyncrhonous task systemsOmri Mor (UIUC)Programming Languages and Runtimes
Runtime Optimization through Explicit Memory ManagementNicolas Denoyelle (ANL)Programming Languages and Runtimes
There and back again ... From lists of irregular data structures to AoSoA and back.Andreas Beckmann (JSC)Programming Languages and Runtimes
Task-Aware MPI (TAMPI) and OpenMPVicenç Beltran (BSC)Programming Languages and Runtimes, Apps and Mini-Apps
ST M3.2Application-aware black-box online power controlValentin Reis (ANL)Programming Languages and Runtimes
Wednesday 10:00
Checkpointing Beyond Resilience: Use Cases and BenefitsBogdan Nicolae (ANL)I/O Storage and In-Situ Processing; Resilience
Range and span of error reporting and recovery for modular MPI fault toleranceAurelien Bouteiller (UTK)Resilience
Reducing the failure overhead by lowering the synchronicity in the local recovery of MPI applicationsNuria Losada (UTK)Resilience
DeepSZ: A Novel Framework to Compress Deep Neural Networks by Using Error-Bounded Lossy CompressionXin Liang (ANL)I/O, Storage and In-Situ Processing
Characterizing and Understanding HPC Job Failures over The 2K-day Life of IBM BlueGene/Q SystemSheng Di (ANL)I/O, Storage and In-Situ Processing
Modeling, Simulation and AI for Optimal checkpointing strategiesKento Sato (RIKEN R-CCS)I/O Storage and In-Situ Processing; Resilience
BOS Organizer Speaker Title
Python in HPCLaxmikant (Sanjay) Kale Laxmikant (Sanjay) KaleCharm4py: Parallel Programming with Python and Charm++
Wednesday 10:00
Laxmikant (Sanjay) KalePrivatization Technique for Process Virtualization: AMPI (+ PiP)
Rosa BadiaPython @ BSC
Robert SpeckParallel-in-Time integration with pySDC: From prototyping to applications
Daniel KatzPython-based workflows with Parsl
Daniel CoquelinThe HeAT Framework
All presentersDiscussion about possible project and collaborations
AutotuningPiotr LuszczekPiotr Luszczek Introduction to Autotuning in JLESC
Monday 13:30
Toshiyuki ImamuraResearch advancement in autotuning in libraries and applications
Sandra CatalanAutotuning approaches
Mike TsaiDistributed Autotuning
FPGAKazutomo (Kaz) YoshiiKazutomo (Kaz) YoshiiOpening
Tuesday 14:00
Rosa BadiaSupporting task creation inside FPGA devices
Leonardo Bautista-GomezExperimental evaluation of supply voltage underscaling in FPGAs
Eric RuttenAutonomic Management of Reconfigurations in DPR FPGA-based Systems
All presentersOpen discussions (platforms, programming languages, benchmarking, cultivating FPGA-HPC community)
Kazutomo (Kaz) YoshiiWrap up
ARMMitsuhisa Sato Tuesday 10:00
ARMMitsuhisa Sato Tuesday 13:30
Title Topic Presenter Authors Abstract
Performance and Power Consumption Analysis of Arm Scalable Vector Extension with gem5 SimulatorPerformance ToolsTetsuya OdajimaTetsuya Odajima, Yuetsu Kodama, Miwako Tsuji, Mitsuhisa Sato (RIKEN)
SLATE: Software for Linear Algebra Targeting ExascaleNumerical MethodsJakub KurzakJakub Kurzak, University of Tennessee Mark Gates, University of Tennessee Ali Charara, University of Tennessee Asim YarKhan, University of Tennessee Jack Dongarra, University of Tennessee
BONSAI: Benchmarking OpeN Software Autotuning InfrastructurePerformance ToolsJakub KurzakJakub Kurzak, University of Tennessee Mike Tsai, University of Tennessee Mark Gates, University of Tennessee Jack Dongarra, University of Tennessee
PaRSEC - a data-flow task-based runtimeProgramming Languages and RuntimesYu Pei, Qinglei CaoReazul Hoque, Yu Pei, Qinglei Cao - University of Tennessee, Knoxville
Extending Open MPI with Tool and Resilience SupportParallel Programming models and runtime; Performance tools; ResilienceDavid Eberius, Dong ZhongDavid Eberius, Dong Zhong (Innovative Computing Laboratory, University of Tennessee)
PRIONN: Predicting Runtime and IO using Neural NetworksI/O, Storage and In-Situ ProcessingMichael WyattMichael Wyatt (UTK), Stephen Herbein (LLNL), Todd Gamblin (LLNL), Adam Moody (LLNL), Dong H Ahn (LLNL), Michela Taufer (UTK)
Modeling Record-and-Replay for Nondeterministic Applications on Exascale SystemsResilience, Performance tools, Applications and mini-appsDylan ChappDylan Chapp (University of Delaware, University of Tennessee, Knoxville), Danny Rorabaugh (University of Tennessee, Knoxville), Michela Taufer (University of Tennesee, Knoxville)
Performance improvements in Open MPIProgramming Languages and RuntimesXi Luo, Thananon PatinyasakdikulXi Luo, Thananon Patinyasakdikul (UTK)
Pseudo-Assembly Programming for Batched Matrix Factorizationscientific simulation, high-order PDE methodsMike Tsai (UTK)Mike Tsai (UTK) Piotr Luszczek (UTK) Jakub Kurzak (UTK) Jack Dongarra (UTK)
Improve Exascale IO via an Adaptive Lossy CompressorBig Data, I/O and in-situ visualizationXin LiangXin Liang (UCR), Sheng Di (ANL), Sihuan Li (UCR), Dingwen Tao (UA), Bogdan Nicolae (ANL), Zizhong Chen (UCR), Franck Cappello (ANL)
Towards SDC Resilient Error Bounded Lossy CompressorResilience, ABFT (Algorithm Based Fault Tolerance), Lossy compressionSihuan LiSihuan Li, UC Riverside; Sheng Di, ANL; Xin Liang, UC Riverside; Zizhong Chen, UC Riverside; Franck Cappello, ANL
Towards Unified Tasking on CPUs and GPUsParallel Programming ModelsLaura MorgensternLaura Morgenstern (JSC), Ivo Kabadshow (JSC)
Holistic Measurement Driven System AssessmentArchitecture, ResilienceWilliam KramerPI: Bill Kramer (NCSA/UIUC) NCSA: Greg Bauer Brett Bode Jeremy Enos Aaron Saxton Mike Showerman UIUC: Saurabh Jha (CS/CSL) Ravi Iyer (ECE/CSL) Zbigniew Kalbarczyk (ECE/CSL) SNL: Jim Brandt Ann Gentile
Modeling HPC applications for in situ AnalyticsI/O, Storage and In-Situ ProcessingValentin HONOREGuillaume AUPY, Brice GOGLIN, Valentin HONORE (Inria, LaBRI, Univ. Bordeaux, 33400 Talence, France) - Bruno RAFFIN (Univ. Grenoble Alpes, Inria, CNRS, Grenoble INP, LIG, 38000 Grenoble, France)
Enhancing Deep Learning towards Exascale with the DEEP-EST Modular Supercomputer ArchitectureParallel Programming models and runtime; Performance tools;Sedona RoccoErnir Erlingsson, Gabriele Cavallaro, Morris Riedel, Rocco Sedona and Helmut Neukirchen (JSC)

Plenary Talks

Title Presenter
Not Your Grandfather’s Tractor – How AI & IoT Are Transforming Production AgricultureMark Moran
John Deere
Abstract:

After some background on the sweeping changes happening in agricultural technology as a result of General Purpose Technologies, Mark will discuss how AI and IoT are merging on the farm, creating some compelling problems to be solved on the edge.

Programming workflows for Advanced Cyberinfrastructure PlatformsRosa M. Badia
Barcelona Supercomputing Center
Abstract:

In the design of an advanced cyberinfrastructure platform (ACP), that involve sensors, edge devices, instruments, computing power in the cloud and HPC, a key aspect is how to describe the applications to be executed in such platform.

Very often these applications are not standalone, but involve a set of sub-applications or steps composing a workflow. The scientists then rely on effective environments to describe their workflows and engines to manage them in complex infrastructures.

COMPSs is a task-based programming model that enables the development of workflows that can be executed in parallel in distributed computing platform. The workflows that we currently support may involve different types of tasks, such as parallel simulations (MPI) or analytics (i.e., written in Python thanks to PyCOMPSs, the Python binding for COMPSs). COMPSs, through and storage interface, makes transparent the access to persistent data stored in key-value databases (Hecuba) or object-oriented distributed storage environments (dataClay).

While COMPSs has been developed from its early times for highly distributed environments, we have been extending it to deal with more challenging environments, with edge devices and components in the fog, that can appear and disappear.

Delivering on the Exascale Computing Project Mission for the U.S. Department of EnergyDouglas B. (Doug) Kothe
Oak ridge National Laboratory
Abstract:

The vision of the U.S. Department of Energy (DOE) Exascale Computing Project (ECP), initiated in 2016 as a formal DOE project executing through 2023, is to accelerate innovation with exascale simulation and data science solutions that enhance U.S. economic competitiveness, strengthen our national security, and change our quality of life. ECP’s mission is to deliver exascale-ready applications and solutions that address currently intractable problems of strategic importance and national interest; create and deploy an expanded and vertically integrated software stack on DOE HPC exascale and pre-exascale systems, thereby defining the enduring US exascale ecosystem; and leverage U.S. HPC vendor research activities and products into DOE HPC exascale systems. The project is a joint effort of two DOE programs: the Office of Science Advanced Scientific Computing Research Program and the National Nuclear Security Administration Advanced Simulation and Computing Program. ECP’s RD&D activities are carried out by over 100 teams of scientists and engineers from the DOE national laboratories, universities, and U.S. industry. These teams have been working together since the fall of 2016 on the development of applications, software technologies, and hardware technologies and architectures:

Applications: Creating or enhancing the predictive capability of applications through algorithmic and software advances via co-design centers; targeted development of requirements-based models, algorithms, and methods; systematic improvement of exascale system readiness and utilization; and demonstration and assessment of effective software integration.

Software Technologies: Developing and delivering a vertically integrated software stack containing advanced mathematical libraries, extreme-scale programming environments, development tools, visualization libraries, and the software infrastructure to support large-scale data management and data science for science and security applications.

Hardware and Integration: Supporting U.S. HPC vendor R&D focused on innovative architectures for competitive exascale system designs; objectively evaluating hardware designs; deploying an integrated and continuously tested exascale software ecosystem at DOE HPC facilities; accelerating application readiness on targeted exascale architectures; and training on key ECP technologies to accelerate the software development cycle and optimize productivity of application and software developers.

Illustrative examples will be given on how the ECP teams are delivering in these three areas of technical focus, with specific emphasis on recent work on the world’s #1 supercomputer, namely the Summit leadership system at Oak Ridge National Laboratory.

Project Talks

Title Presenter Project Abstract
PT M1.1Comparing the performance of rigid, moldable and grid-shaped applications on failure-prone HPC platformsThomas HeraultOptimization of Fault-Tolerance Strategies for Workflow Applications
Monday 10:30
Using Lossy Compression for Linear Solver ResilienceMirco AltenberndEffective Use of Lossy Compression for Numerical Linear Algebra Resilience and Performance
Exploring the Variability and Error Distributions of Lossy Compression AlgorithmsJon Calhoun (Clemson University)Checkpoint/Restart of/from lossy state
Reducing the memory footprint in Krylov methods through lossy compressionNick Schenkels (Inria) Checkpoint/Restart of/from lossy state
PT M1.2Review on standard eigensolvers on a high-end GPU systemToshiyuki Imamura (RIKEN)HPC libraries for solving dense symmetric eigenvalue problems
Monday 10:30
Adding Inter-Node Communication to a C++ Tasking Framework for Synchronization-Critical ApplicationsIvo Kabadshow (JSC)Scalability Enhancements to FMM for Molecular Dynamics Simulations
Fast Integrators for Scalable Quantum Molecular DynamicsAlina Kononov (UIUC)Fast Integrators for Scalable Quantum Molecular Dynamics
Solving the Bethe-Salpeter eigenvalue problem on thousands of coresXiao Zhang (University of Illinois)The ChASE library for large Hermitian eigenvalue problems
PT M1.3Evaluating OmpSs2 Runtime for XMP 2.0 PGAS Task ModelJinpil Lee (RIKEN R-CCS)Sharing and extension of OmpSs2 runtime for XMP 2.0 PGAS task model
Monday 10:30
Analyses and Transformations for Reverse Mode Differentiation of OpenMPJan Hueckelheim (Imperial College London / Argonne)Shared Infrastructure for Source Transformation Automatic Differentiation
Performance of S-Step and Pipelined Krylov MethodsIchitaro Yamazaki (UTK)Reducing Communication in Sparse Iterative and Direct Solvers
Designing Scalable Solvers for Enlarged Krylov Subspace MethodsShelby Lockhart (University of Illinois)Reducing Communication in Sparse Iterative and Direct Solvers
PT A1.2Project update: Developer tools for porting and tuning parallel applications on extreme-scale parallel systemsBrian Wylie (JSC)Developer tools for porting and tuning parallel applications on extreme-scale parallel systems
Monday 13:30
On the Applicability of PEBS based Online Memory Access Tracking for Heterogeneous Memory Management at ScaleLeonardo Bautista (BSC)Deep Memory Hierarchies
Monitoring MPI and forecasting network usageBogdan Nicolae (ANL)Towards accurate network utilization forecasting using portable MPI-level monitoring
Preliminary Result of MPI International SurveyAtsushi Hori (Riken)MPI International Survey
PT A3.1Rescaling Transient Distributed Storage SystemsNathanaël CheriereReconfiguring Distributed Storage Systems on HPC infrastructures
Wednesday 13:30
Experiment Management and Repeatability Services in ChameleonKate KeaheyAdvancing Chameleon and Grid'5000 testbeds
Improving the Performance and Energy Efficiency of HPC Applications Using Autonomic Computing TechniquesEric Rutten, InriaImproving the Performance and Energy Efficiency of HPC Applications Using Autonomic Computing Techniques
Leveraging Blob-Based Storage for Stream ProcessingPierre Matri, ANLTowards Blob-Based Convergence Between HPC and Big Data
PT A3.2Heterogeneous Hierarchical Workflow CompositionOrçun YILDIZ (ANL)Extreme-Scale Workflow Tools - Swift, Decaf, Damaris, FlowVR
Wednesday 13:30
Evaluating OpenCL Kernels for FPGA platformsZheming JinEvaluating high-level programming models for FPGA platforms
Cultivating a community about FPGA-HPC platformsKazutomo Yoshii (ANL)Evaluating high-level programming models for FPGA platforms
Status of the Simplified Sustained System performance benchmark ProjectMiwako Tsuji (RIKEN)Simplified Sustained System performance benchmark
Toward MPI-based Workflow Execution: Applications and StandardizationJustin M. Wozniak (ANL)Resource Management, Scheduling, and Fault-Tolerance for HPC Workflows

Short Talks

Title Presenter Topic
ST A1.1Scheduling stochastic tasks under buget/deadline constraintsYves RobertProgramming Languages and Runtimes
Monday 15:30
Tasking Meets GPUs: Fighting Deadlocks and Other MonstersLaura Morgenstern, JSCNumerical Methods, Programming Languages and Runtimes
Python-based workflows & software sustainabilityDaniel S. Katz (UIUC)Programming Languages and Runtimes
Scheduling stochastic jobs on HPC platformsHongyang Sun (INRIA)Programming Languages and Runtimes
Runtime Systems to Support GPU Acceleration of Irregular ProblemsThomas Herault (UTK)Performance Tools 2 Programming Languages and Runtimes
Scalable Dynamic DAG Building for data-flow task-based runtime systemsReazul Hoque (UTK)Programming Languages and Runtimes
Integrating Futures as Async Provider into PaRSEC Runtime SystemYu Pei (UTK)Programming Languages and Runtimes
ST A1.2Evaluating High-throughput Applications Models for in situ AnalyticsValentin Honoré (Univ. Bordeaux/Inria)I/O, Storage and In-Situ Processing
Monday 15:30
Scalable Data Ingestion for Stream ProcessingGabriel AntoniuI/O, Storage and In-Situ Processing
Combining Checkpointing and Replication for Reliable Execution of Linear WorkflowsValentin Le Fevre (Inria)Resilience
Optimal algorithms for Adjoint ComputationJulien Herrmann (Inria Bordeaux)I/O, Storage and In-Situ Processing
Analytics4MD: In-Situ Data Analytics for Next Generation Molecular Dynamics Workflows.Stephen Thomas (UTK)I/O, Storage and In-Situ Processing, Apps & Mini-Apps
Analytics4MD: In-Situ Data Analytics for Next Generation Molecular Dynamics Workflows.Stephen Thomas (UTK)I/O, Storage and In-Situ Processing, Apps & Mini-Apps
Developing Holistic Measurement Driven System Intelligent AssessmentBill Kramer (NCSA)Architecture, Resilience
ST A1.3Scalable MPI completionThananon Patinyasakdikul (UTK)Programming Languages and Runtimes
Monday 15:30
Improved Convolution Implementations on NVIDIA GPUsMarc Jorda (BSC)Numerical Methods; Programming Languages and Runtimes
Towards Continuous Benchmarking (CB) for sustainable software developmentHartwig Anzt (UTK)PDE solvers for CFD, data-sparse and sparse formats, GPU asynchronous programming, preconditioned iterative methods with optimal communication
Classification methods for Polar Stratospheric CloudsRocco SedonaNumerical Methods
BONSAI: Benchmarking OpeN Software Autotuning InfrastructureJakub Kurzak (UTK)Performance Tools
The HeAT FrameworkDaniel Gage Coquelin (JSC)Advanced Architectures
DeepHyper: Scalable Asynchronous Neural Architecture and Hyperparameter Search for Deep Neural NetworksRomit Maulik (ANL)Machine Learning
ST M2.1HPC Computations for Respiratory Flows - Towards Personalized MedicineAndreas Lintermann (JSC)Numerical Methods
Tuesday 10:00
MagmaDNN - High-Performance Data Analytics for Manycore GPUs and CPUsKwai Wong (UTK)Numerical Methods
SLATE: Software for Linear Algebra Targeting ExascaleJakub Kurzak (UTK)Numerical Methods
Mixed-precision solversAzzam Haidar (NVidia)Numerical Methods
ParILUT - A Parallel Threshold ILU for Manycore ParallelismHartwig Anzt (UTK)PDE solvers for CFD, data-sparse and sparse formats, GPU asynchronous programming, preconditioned iterative methods with optimal communication
Adaptive Precision PreconditioningHartwig Anzt (UTK)PDE solvers for CFD, data-sparse and sparse formats, Adaptive precision Preconditioning
Mesh partitioning and load balancing on heterogeneous systemsRicard Borrell (BSC)Numerical Methods
ST M2.2Characterization of Power and Performance in Data-Intensive Applications using MapReduce over MPIMichela Taufer (UTK)Apps & Mini-Apps; I/O, Storage and In-Situ Processing
Tuesday 10:00
POP2 - The Quest for Performance ContinuesBernd Mohr (JSC)Performance Tools
Understanding native event semanticsAnthony Danalis (UTK)Performance Tools
Using Software-based Performance Counters to Analyze MPI PerformanceDavid Eberius (UTK)Performance tools
A design space exploration targeting next-generation HPC architecturesConstantino Gómez (BSC)Advanced Architecture
Exploring Next Generation High Performance Computing Architecture for Post-Moore EraMasaaki Kondo (RIKEN-CCS)Advanced Architectures, Performance Tools
ST A2.2High-Performance Custom Computing System with Stratix10 FPGAsKentaro Sano, Riken R-CCSAdvanced Architectures
Tuesday 13:30
Dynamic Resource Partitioning for Complex Workloads using ContainersSwann Perarnau (ANL)Advanced Architectures, Performance Tools, Apps & Mini-apps
Improving the performance of a AI-based Big Data application using Edge computingPedro Silva (INSA de Rennes / Inria)Advanced Architectures
The Load Balancing SandboxVincent ReverdyParallel Programming models and runtime, Applications and mini-apps
Downscaling Soil Moisture Data from Remote SensingDanny Rorabaugh (UTK)Apps & Mini-Apps
FFT for large-scale heterogeneous systemsAlan Ayala (UTK)Apps & Mini-Apps
Scalable graph algorithms for cluster-findingLaxmikant Kale and Raghavendra KanakagiriApps & Mini-Apps
Porting Applications towards Modular SupercomputingAlexander Sebastian Achilles (JSC)Advanced Architectures, Apps & Mini-Apps
ST M3.1Supporting multiple virtual ranks in MPI: exploration of techniques including process-in-processLaxmikan Kale and Same White (UIUC)Programming Languages and Runtimes
Wednesday 10:00
Performance and Power Consumption Analysis of Arm Scalable Vector Extension with gem5 SimulatorTetsuya Odajima, (RIKEN R-CCS)Performance Tools
On the Impact of OpenMP Task GranularityChristian Perez (Inria)Programming Languages and Runtimes
SLATE-PaRSEC interface, Bridging Two Dataflow RepresentationsDamien Genet (UTK)Programming Languages and Runtimes
Communication runtime for asyncrhonous task systemsOmri Mor (UIUC)Programming Languages and Runtimes
Runtime Optimization through Explicit Memory ManagementNicolas Denoyelle (ANL)Programming Languages and Runtimes
There and back again ... From lists of irregular data structures to AoSoA and back.Andreas Beckmann (JSC)Programming Languages and Runtimes
Task-Aware MPI (TAMPI) and OpenMPVicenç Beltran (BSC)Programming Languages and Runtimes, Apps and Mini-Apps
ST M3.2Application-aware black-box online power controlValentin Reis (ANL)Programming Languages and Runtimes
Wednesday 10:00
Checkpointing Beyond Resilience: Use Cases and BenefitsBogdan Nicolae (ANL)I/O Storage and In-Situ Processing; Resilience
Range and span of error reporting and recovery for modular MPI fault toleranceAurelien Bouteiller (UTK)Resilience
Reducing the failure overhead by lowering the synchronicity in the local recovery of MPI applicationsNuria Losada (UTK)Resilience
DeepSZ: A Novel Framework to Compress Deep Neural Networks by Using Error-Bounded Lossy CompressionXin Liang (ANL)I/O, Storage and In-Situ Processing
Characterizing and Understanding HPC Job Failures over The 2K-day Life of IBM BlueGene/Q SystemSheng Di (ANL)I/O, Storage and In-Situ Processing
Modeling, Simulation and AI for Optimal checkpointing strategiesKento Sato (RIKEN R-CCS)I/O Storage and In-Situ Processing; Resilience

Posters

Title Topic Presenter Authors Abstract
Performance and Power Consumption Analysis of Arm Scalable Vector Extension with gem5 SimulatorPerformance ToolsTetsuya OdajimaTetsuya Odajima, Yuetsu Kodama, Miwako Tsuji, Mitsuhisa Sato (RIKEN)
SLATE: Software for Linear Algebra Targeting ExascaleNumerical MethodsJakub KurzakJakub Kurzak, University of Tennessee Mark Gates, University of Tennessee Ali Charara, University of Tennessee Asim YarKhan, University of Tennessee Jack Dongarra, University of Tennessee
BONSAI: Benchmarking OpeN Software Autotuning InfrastructurePerformance ToolsJakub KurzakJakub Kurzak, University of Tennessee Mike Tsai, University of Tennessee Mark Gates, University of Tennessee Jack Dongarra, University of Tennessee
PaRSEC - a data-flow task-based runtimeProgramming Languages and RuntimesYu Pei, Qinglei CaoReazul Hoque, Yu Pei, Qinglei Cao - University of Tennessee, Knoxville
Extending Open MPI with Tool and Resilience SupportParallel Programming models and runtime; Performance tools; ResilienceDavid Eberius, Dong ZhongDavid Eberius, Dong Zhong (Innovative Computing Laboratory, University of Tennessee)
PRIONN: Predicting Runtime and IO using Neural NetworksI/O, Storage and In-Situ ProcessingMichael WyattMichael Wyatt (UTK), Stephen Herbein (LLNL), Todd Gamblin (LLNL), Adam Moody (LLNL), Dong H Ahn (LLNL), Michela Taufer (UTK)
Modeling Record-and-Replay for Nondeterministic Applications on Exascale SystemsResilience, Performance tools, Applications and mini-appsDylan ChappDylan Chapp (University of Delaware, University of Tennessee, Knoxville), Danny Rorabaugh (University of Tennessee, Knoxville), Michela Taufer (University of Tennesee, Knoxville)
Performance improvements in Open MPIProgramming Languages and RuntimesXi Luo, Thananon PatinyasakdikulXi Luo, Thananon Patinyasakdikul (UTK)
Pseudo-Assembly Programming for Batched Matrix Factorizationscientific simulation, high-order PDE methodsMike Tsai (UTK)Mike Tsai (UTK) Piotr Luszczek (UTK) Jakub Kurzak (UTK) Jack Dongarra (UTK)
Improve Exascale IO via an Adaptive Lossy CompressorBig Data, I/O and in-situ visualizationXin LiangXin Liang (UCR), Sheng Di (ANL), Sihuan Li (UCR), Dingwen Tao (UA), Bogdan Nicolae (ANL), Zizhong Chen (UCR), Franck Cappello (ANL)
Towards SDC Resilient Error Bounded Lossy CompressorResilience, ABFT (Algorithm Based Fault Tolerance), Lossy compressionSihuan LiSihuan Li, UC Riverside; Sheng Di, ANL; Xin Liang, UC Riverside; Zizhong Chen, UC Riverside; Franck Cappello, ANL
Towards Unified Tasking on CPUs and GPUsParallel Programming ModelsLaura MorgensternLaura Morgenstern (JSC), Ivo Kabadshow (JSC)
Holistic Measurement Driven System AssessmentArchitecture, ResilienceWilliam KramerPI: Bill Kramer (NCSA/UIUC) NCSA: Greg Bauer Brett Bode Jeremy Enos Aaron Saxton Mike Showerman UIUC: Saurabh Jha (CS/CSL) Ravi Iyer (ECE/CSL) Zbigniew Kalbarczyk (ECE/CSL) SNL: Jim Brandt Ann Gentile
Modeling HPC applications for in situ AnalyticsI/O, Storage and In-Situ ProcessingValentin HONOREGuillaume AUPY, Brice GOGLIN, Valentin HONORE (Inria, LaBRI, Univ. Bordeaux, 33400 Talence, France) - Bruno RAFFIN (Univ. Grenoble Alpes, Inria, CNRS, Grenoble INP, LIG, 38000 Grenoble, France)
Enhancing Deep Learning towards Exascale with the DEEP-EST Modular Supercomputer ArchitectureParallel Programming models and runtime; Performance tools;Sedona RoccoErnir Erlingsson, Gabriele Cavallaro, Morris Riedel, Rocco Sedona and Helmut Neukirchen (JSC)

Participants 

JLESC9 Group
JLESC9 Group in front of Sunsphere

Name Affiliation Arrival Departure
Andreas Beckmann Jülich Supercomputing Centre 4/14/2019 12:00:00 4/18/2019 11:59:00
Andreas Lintermann Jülich Supercomputing Centre 4/14/2019 20:00:00 4/17/2019 13:45:00
Anne Nikodemus Jülich Supercomputing Centre 4/14/2019 19:00:00 4/16/2019 17:00:00
Bernd Mohr Jülich Supercomputing Centre 4/14/2019 20:00:00 4/18/2019 10:00:00
Brian Wylie Jülich Supercomputing Centre 4/7/2019 21:00:00 4/18/2019 9:00:00
Daniel Coquelin Jülich Supercomputing Centre 4/13/2019 18:00:00 4/17/2019 5:00:00
Ivo Kabadshow Jülich Supercomputing Centre 4/14/2019 1:00:00 4/18/2019 0:00:00
Laura Morgenstern Jülich Supercomputing Centre 4/14/2019 13:00:00 4/18/2019 12:00:00
Mirco Altenbernd Jülich Supercomputing Centre 4/14/2019 18:00:00 4/17/2019 18:00:00
Norbert Attig Jülich Supercomputing Centre 4/13/2019 20:00:00 4/17/2019 13:00:00
Ruth Schoebel Jülich Supercomputing Centre 4/13/2019 22:00:00 4/18/2019 5:00:00
Robert Speck Jülich Supercomputing Centre 4/13/2019 20:00:00 4/17/2019 13:00:00
Rocco Sedona Jülich Supercomputing Centre 4/14/2019 19:00:00 4/18/2019 15:45:00
Thomas Lippert Jülich Supercomputing Centre 4/14/2019 23:45:00 4/18/2019 14:50:00
Wolfgang Frings Jülich Supercomputing Centre 4/14/2019 19:00:00 4/17/2019 15:00:00
Anthony Danalis The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Ana Gainaru The University of Tennessee 4/14/2019 2:00:00 4/17/2019 22:00:00
George Bosilca The University of Tennessee 4/14/2019 9:00:00 4/17/2019 18:00:00
Aurelien Bouteiller The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Ali Charara The University of Tennessee 4/15/2019 8:00:00 4/17/2019 17:00:00
Dylan Chap The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
David Eberius The University of Tennessee 4/15/2019 8:00:00 4/17/2019 16:00:00
Damien Genet The University of Tennessee 4/14/2019 18:00:00 4/18/2019 11:00:00
Jack Dongarra The University of Tennessee 4/15/2019 8:00:00 4/17/2019 15:00:00
Danny Rorabaugh The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Dong Zhong The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Hartwig Anzt The University of Tennessee 4/15/2019 8:00:00 4/17/2019 15:00:00
Thomas Herault The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Ichitaro Yamazaki The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Jiali Li The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Jon Calhoun The University of Tennessee 4/14/2019 18:00:00 4/17/2019 17:00:00
Joseph Teague The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Jakub Kurzak The University of Tennessee 4/15/2019 8:00:00 4/17/2019 17:00:00
Piotr Luszczek The University of Tennessee 4/15/2019 0:01:00 4/18/2019 23:59:00
Michael Wyatt The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Nuria Losada Lopez Valcarcel The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Qinglei Cao The University of Tennessee 4/15/2019 8:30:00 4/17/2019 15:00:00
Gerald Ragghianti The University of Tennessee 4/14/2019 1:00:00 4/18/2019 1:00:00
Reazul Hoque The University of Tennessee 4/15/2019 2:00:00 4/17/2019 10:00:00
Ryan Marshall The University of Tennessee 4/14/2019 17:30:00 4/17/2019 15:30:00
Stephen Thomas The University of Tennessee 4/15/2019 9:00:00 4/15/2019 15:00:00
Michela Taufer The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Terry Moore The University of Tennessee 4/15/2019 8:00:00 4/17/2019 15:00:00
Stanimire Tomov The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Anthony Skjellum The University of Tennessee 4/15/2019 18:00:00 4/17/2019 15:00:00
Thananon Patinyasakdikul The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Xi Luo The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Asim YarKhan The University of Tennessee 4/15/0009 9:00:00 4/17/2019 23:59:00
Yu Pei The University of Tennessee 4/15/2019 8:00:00 4/17/2019 15:00:00
Yaohung Tsai The University of Tennessee 4/15/2019 9:00:00 4/17/2019 15:00:00
Azzam Haidar The University of Tennessee 4/15/2019 08:00:00 4/17/2019 15:00:00
Atsushi Hori RIKEN R-CCS 4/14/2019 16:00:00 4/18/2019 6:00:00
Toshiyuki Imamura RIKEN R-CCS 4/14/2019 21:00:00 4/18/2019 0:00:00
Jinpil Lee RIKEN R-CCS 4/14/2019 12:00:00 4/18/2019 10:00:00
Kentaro Sano RIKEN R-CCS 4/14/2019 15:00:00 4/17/2019 16:10:00
Kento Sato RIKEN R-CCS 4/15/2019 20:35:00 4/18/2019 8:15:00
Masaaki Kondo RIKEN R-CCS 4/14/2019 15:00:00 4/16/2019 15:00:00
Satoshi Matsuoka RIKEN R-CCS 4/15/2019 20:35:00 4/17/2019 13:10:00
Miwako Tsuji RIKEN R-CCS 4/15/2019 15:00:00 4/17/2019 15:00:00
Mitsuhisa Sato RIKEN R-CCS 4/15/2019 17:00:00 4/17/2019 15:00:00
Takahiro Ogura RIKEN R-CCS 4/14/2019 14:20:00 4/18/2019 8:15:00
Tetsuya Odajima RIKEN R-CCS 4/14/2019 20:35:00 4/17/2019 18:40:00
Yuetsu Kodama RIKEN R-CCS 4/15/2019 21:30:00 4/17/2019 15:00:00
Brendan McGinty The University of Illinois 4/15/2019 07:30:00 4/16/2019 15:00:00
Brett Bode The University of Illinois 4/14/2019 17:00:00 4/18/2019 9:00:00
Daniel Katz The University of Illinois 4/14/2019 22:00:00 4/18/2019 12:00:00
Colleen Heinemann The University of Illinois 4/14/2019 17:00:00 4/18/2019 9:00:00
Iftekhar Ahmed The University of Illinois 4/14/2019 12:00:00 4/17/2019 11:00:00
Laxmikant Kale The University of Illinois 4/16/2019 12:00:00 4/17/2019 14:00:00
Alina Kononov The University of Illinois 4/14/2019 15:00:00 4/18/2019 10:00:00
Scott Poole The University of Illinois 4/15/2019 17:00:00 4/17/2019 15:00:00
Mike Showerman The University of Illinois 4/14/2019 16:00:00 4/18/2019 12:00:00
Omri Mor The University of Illinois 4/15/2019 16:00:00 4/17/2019 17:00:00
Raghavendra Kanakagiri The University of Illinois 4/16/2019 23:00:00 4/17/2019 15:00:00
Aaron Saxton The University of Illinois 4/14/2019 12:00:00 4/18/2019 12:00:00
Robert Sisneros The University of Illinois 4/14/2019 16:00:00 4/17/2019 16:00:00
Shelby Lockhart The University of Illinois 4/14/2019 0:00:00 4/18/2019 0:00:00
Marc Snir The University of Illinois 4/14/2019 15:00:00 4/16/2019 18:00:00
William Gropp The University of Illinois 4/14/2019 18:00:00 4/17/2019 8:00:00
William Kramer The University of Illinois 4/14/2019 20:00:00 4/17/2019 17:00:00
Xiao Zhang The University of Illinois 4/14/2019 17:00:00 4/17/2019 10:00:00
Vincent Reverdy The University of Illinois 4/15/2019 17:00:00 4/17/2019 15:00:00
Bogdan Nicolae Argonne National Laboratory 4/13/2019 20:35:00 4/17/2019 17:20:00
Franck Cappello Argonne National Laboratory 4/15/2019 17:00:00 4/18/2019 10:00:00
Kazutomo Yoshi Argonne National Laboratory 4/16/2019 11:00:00 4/18/2019 10:00:00
Nicolas Denoyelle Argonne National Laboratory 4/16/2019 11:00:00 4/17/2019 17:00:00
Orçun Yildiz Argonne National Laboratory 4/15/2019 17:00:00 4/18/2019 9:00:00
Pierre Matri Argonne National Laboratory 4/14/2019 18:00:00 4/18/2019 8:00:00
Romit Maulik Argonne National Laboratory 4/15/2019 10:00:00 4/17/2019 13:00:00
Sheng Di Argonne National Laboratory 4/16/2019 11:00:00 4/17/2019 17:00:00
Sihuan Li Argonne National Laboratory 4/16/2019 10:00:00 4/17/2019 15:30:00
Swann Perarnau Argonne National Laboratory 4/15/2019 23:00:00 4/18/2019 9:00:00
Valentin Reis Argonne National Laboratory 4/15/2019 23:50:00 4/18/2019 3:00:00
Valerie Taylor Argonne National Laboratory 4/15/2019 23:00:00 4/16/2019 8:00:00
Justin Wozniak Argonne National Laboratory 4/14/2019 18:00:00 4/17/2019 15:00:00
Xin Liang Argonne National Laboratory 4/16/2019 10:00:00 4/17/2019 15:30:00
Zheming Jin Argonne National Laboratory 4/15/2019 11:00:00 4/18/2019 11:59:00
Christian Perez INRIA 4/14/2019 20:00:00 4/17/2019 17:00:00
Christine Morin INRIA 4/14/2019 22:00:00 4/18/2019 12:00:00
Eric Rutten INRIA 4/15/2019 23:45:00 4/17/2019 15:00:00
Gabriel Antoniu INRIA 4/14/2019 20:00:00 4/17/2019 17:00:00
Hongyang Sun INRIA 4/13/2019 17:00:00 4/18/2019 11:00:00
Julien Herrmann INRIA 4/14/2019 17:00:00 4/17/2019 18:00:00
Nathanaël Cheriere INRIA 4/13/2019 22:00:00 4/18/2019 15:00:00
Nick Schenkels INRIA 4/14/2019 20:00:00 4/18/2019 10:00:00
Pedro Paulo de Souza Bento da Silva INRIA 4/13/2019 18:44:00 4/17/2019 17:00:00
Valentin Honoré INRIA 4/13/2019 19:00:00 4/17/2019 17:00:00
Valentin Le Fèvre INRIA 4/13/2019 21:00:00 4/18/2019 14:00:00
Yves Robert INRIA 4/13/2019 15:00:00 4/18/2019 10:00:00
Constantino Gómez Barcelona Supercomputing Center 4/14/2019 15:00:00 4/17/2019 17:00:00
Germán Llort Barcelona Supercomputing Center 4/13/2019 9:00:00 4/18/2019 15:00:00
judit gimenez Barcelona Supercomputing Center 4/13/2019 0:00:00 4/17/2019 12:00:00
Leonardo Bautista Gomez Barcelona Supercomputing Center 4/15/2019 13:00:00 4/17/2019 13:00:00
Marc Jorda Barcelona Supercomputing Center 4/14/2019 16:00:00 4/18/2019 12:00:00
Jesus Labarta Barcelona Supercomputing Center 4/14/2019 19:47:00 4/17/2019 15:25:00
Ricard Borrell Pol Barcelona Supercomputing Center 4/14/2019 0:00:00 4/17/2019 19:00:00
Rosa Badia Barcelona Supercomputing Center 4/15/2019 18:30:00 4/18/2019 13:00:00
Sandra Catalán Barcelona Supercomputing Center 4/13/2019 17:00:00 4/19/2019 12:00:00
Vicenc Beltran Barcelona Supercomputing Center 4/14/2019 17:00:00 4/17/2019 19:00:00
Mark Moran John Deere 4/15/2019 4/16/2019 15:00:00

Organizers 

InriaNCSAUniversity of IllinoisArgonne National LaboratoryBarcelona Supercompting CenterForschungszentrum JülichRIKEN Center for Computational ScienceInnovative Computing LaboratoryUniversity of Tennessee