Conference Agenda

Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).

Session Overview
Date: Wednesday, 11/Sep/2019
9:00 - 10:00Keynote Talk 3
Session Chair: Frans Peters

Ian Foster: Coding the Continuum

Hall S3 
10:00 - 11:00A3: Economics
Session Chair: Carl Kesselman
Hall S4 

Cloud vs On-Premise HPC: a model for comprehensive cost assessment

Luigi Santangelo, Marco Ferretti

University of Pavia, Italy

A Scalable Approach to Econometric Inference

Philip Conrad Nadler, Rossella Arcucci, Yike Guo

Data Science Institute, Imperial College London, United Kingdom

10:00 - 11:00B3: Stream Processing
Session Chair: Thomas Huckle
Hall S1 

Seamless Parallelism Management for Multi-core Stream Processing

Adriano Vogel1, Dalvan Griebler1, Marco Danelutto2, Luiz G. Fernandes1

1Pontifical Catholic University of Rio Grande do Sul, Brazil; 2University of Pisa, Italy

High-Level Stream Parallelism Abstractions with SPar Targeting GPUs

Dinei A. Rockenbach1, Dalvan Griebler1, Marco Danelutto2, Luiz G. Fernandes1

1School of Technology, Pontifical Catholic University of Rio Grande do Sul (PUCRS), Brazil; 2Computer Science Department, University of Pisa, Italy

10:00 - 11:00Symp4-1: ELPA
Hall S10 

ELPA - recent optimizations and library improvements

Pavel Kůs, Andreas Marek, Hermann Lederer

Max Planck Computing and Data Facility, Germany

Recent Advancements in the ELPA Library: Best Practices in Real Applications

Hagen-Henrik Kowalski1, Simone Köcher2, Christoph Scheurer2, Christian Carbogno1

1Fritz Haber Institute of the Max Planck Society, Berlin, Germany; 2Department of Theoretical Chemistry, Technical University of Munich, Munich, Germany

11:00 - 11:30Coffee Break & Exhibition
11:30 - 12:30A4: Deep Learning Applications
Session Chair: Ian Foster
Hall S4 

Deep Generative Model driven Protein Folding Simulations

Heng Ma1, Debsindhu Bhowmik1, Hyungro Lee2, Matteo Turilli2, Michael Young1, Shantenu Jha2,3, Arvind Ramanathan4

1Computational Science and Engineering, Oak Ridge National Laboratory, Oak Ridge, TN 37830, USA; 2RADICAL, ECE, Rutgers University, Piscataway, NJ 08854, USA; 3Brookhaven National Laboratory, Upton, NY 11973, USA; 4Data Science and Learning Division, Argonne National Laboratory, Lemont , IL 60439, USA

First experiences on applying deep learning techniques to prostate cancer detection

Eduardo José Gómez-Hernández, José Manuel García

University of Murcia, Spain

11:30 - 12:30B4: Performance
Session Chair: Siavash Ghiasvand
Hall S1 

AITuning: Machine Learning-based Tuning Tool for Run-Time Communication Libraries

Alessandro Fanfarillo, Davide Del Vento

NCAR, United States of America

Towards benchmarking the asynchronous progress of non-blocking MPI point-to-point and collective operations

Alexey Medvedev

Institute of mechanics, Lomonosov Moscow State University, Russian Federation

11:30 - 12:30C4: Parallel Algorithms
Session Chair: Erik D'Hollander
Hall S8 

On Superlinear Speedups of a Parallel NFA Induction Algorithm

Tomasz Jastrząb

Silesian University of Technology, Poland

A Domain Decomposition Reduced Order Model with Data Assimilation (DD-RODA)

Rossella Arcucci1, Dunhui Xiao1,2,3, Fangxin Fang1,3, Christopher Pain1,3, Yike Guo1

1Data Science Institute, Imperial College London; 2Swansea University, United Kingdom; 3Department of Earth Science and Engineering, Imperial College London

11:30 - 12:30Symp4-2: ELPA
Hall S10 

Efficient Computation of Eigenvalues and Eigenvectors for the Banded Generalized Eigenvalue Problem

Michael Rippl

Technical University of Munich, Germany

Cannon-type triangular matrix multiplication for generalized eigenvalue problems

Bruno Lang1, Valeriy Manin2

1University of Wuppertal; 2University of Wuppertal

12:30 - 14:00Lunch
14:00 - 15:00A5: Power Management
Session Chair: Vladimir Getov
Hall S4 

Dynamic Runtime and Energy Optimization for Power-capped HPC Applications

Bo Wang, Christian Terboven, Matthias Mueller

RWTH Aachen University, Germany

Acceleration of Interactive Multiple Precision Arithmetic Toolbox MuPAT by FMA, SIMD, and OpenMP

Hotaka Yagi1, Emiko Ishiwata1, Hidehiko Hasegawa2

1Tokyo University of Science, Japan; 2Tsukuba University, Japan

14:00 - 15:00B5: Parallel Programming
Session Chair: Arvind Ramanathan
Hall S1 

Unified generation of DG-Kernels for different HPC frameworks

Jan Hönig1, Marcel Koch2, Ulrich Rüde1, Christian Engwer2, Harald Köstler1

1Friedrich-Alexander University Erlangen-Nürnberg; 2University of Münster

Exploiting mathematical properties for higher-order imperative programming

Lutz Schubert, Athanasios Tsitsipas

University of Ulm, Germany

14:00 - 15:00C5: Load Balancing Methods
Session Chair: Bruno Lang
Hall S8 

Progressive load balancing of iterative asynchronous algorithms in distributed memory

Justs Zarins1, Michele Weiland2

1University of Edinburgh, United Kingdom; 2EPCC

Learning-Based Load Balancing for Massively Parallel Simulations of Hot Fusion Plasmas

Theresa Pollinger, Dirk Pflüger

Universität Stuttgart, Germany

15:00 - 15:30Coffee Break & Exhibition
15:30 - 16:30Keynote Talk 4
Session Chair: Marco Danelutto

Jaroslav Cervinka

Hall S3 
16:30 - 17:30Keynote Talk 5
Session Chair: Ludek Kucera

Mikhail Dyakonov

Hall S3