PPAM'2017

PCz

UMCS

Intel

Lenovo

IEEE

Springer

Patronat

Program

download pdf version

Sunday, Sept.10
From 10:30	Registration
From 12:00	Tutorial: Scientific Computing with GPUs - Dominik Goeddeke and Robert Strzodka
From 14:00	Tutorials: Advanced OpenMP Tutorial - Dirk Schmidl Parallel Computing in Java with PCJ library - Piotr Bała team
19:30	Concert and welcome reception in ACK UMCS „Chatka Żaka”, Radziszewskiego 16

Monday, Sept. 11
From 8:40	Registration
8:45 – 9:10	Opening
9:10 – 10:30	Invited talks (2 slots x 40 min.) Chairperson: Boleslaw Szymanski An Overview of High-Performance and a Look at Energy Saving on the Intel Knights Landing for Linear Algebra Computations - Jack Dongarra FLOPS to BYTES: Accelerating Beyond Moore's Law is Data-Oriented - Satoshi Matsuoka
10:30 – 11:00	Coffee break
11:00 – 12:40	Contributed papers in parallel (4 slots x 25 min.) Track A: WS on Power and Energy Aspects of Computations (PEAC) Chairpersons: Laurent Lefevre and Ariel Oleksiak Invited talk: HPC and Energy Efficiency, an oxymoron? - Jean-Marc Pierson Thermal and power-aware fans management in M2DC system - Wojciech Piatek, Ariel Oleksiak, Micha Vor Dem Berge, Jens Hagemeyer and Emmanuel Senechal Efficiency analysis of Intel, AMD and Nvidia 64-bit hardware for memory-bound problems: a case study of ab initio calculations with VASP - Vladimir Stegailov and Vyacheslav Vecher Track B: Main Track: Numerical Algorithms and Parallel Scientific Computing Chairperson: Carl Christian Kjelgaard Mikkelsen Advances in Incremental PCA Algorithms - Tal Halpern and Sivan Toledo The algorithms for forward and backward solution of the Fokker-Planck equation in the heliospheric transport of cosmic rays - Anna Wawrzynczak-Szaban, Renata Modzelewska-Łagodzin and Agnieszka Gil-Świderska Efficient Evaluation of Matrix Polynomials - Niv Hoffman, Oded Schwartz and Sivan Toledo Track C: Main Track: Particle Methods in Simulations Chairperson: Peter Arbenz Fast DEM collision checks on multicore nodes - Konstantinos Krestenitis and Tobias Weinzierl A Space and Bandwidth Efficient Multicore Algorithm for the Particle-in-Cell Method - Yann Barsamian, Arthur Charguéraud and Alain Ketterlin Load Balancing for Particle-in-Cell Plasma Simulation on Multicore Systems - Anton Larin, Sergey Bastrakov, Aleksei Bashinov, Evgeny Efimenko, Igor Surmin, Arkady Gonoskov and Iosif Meyerov The impact of particle sorting on Particle-In-Cell simulation performance - Andrzej Dorobisz, Michał Kotwica, Jacek Niemiec, Oleh Kobzar, Artem Bohdan and Kazimierz Wiatr Track D: WS on Models, Algorithms and Methodologies for Hybrid Parallelism in new HPC Systems Chairpersons: Marco Lapegna, Giuliano Laccetti and Raffaele Montella An experience report on (auto-)tuning of mesh-based PDE solvers on shared memory systems - Dominic Etienne Charrier and Tobias Weinzierl Using GPGPU accelerated interpolation algorithms for marine bathymetry processing with on-premises and cloud based computational resources - Livia Marcellino, Raffaele Montella, Sokol Kosta, Ardelio Galletti, Diana Di Luccio, Vincenzo Santopietro, Mario Ruggieri, Marco Lapegna, Luisa D'Amore and Giuliano Laccetti Relaxing the correctness conditions on concurrent data structures for multicore CPUs. A numerical case study - Giuliano Laccetti, Marco Lapegna, Valeria Mele and Raffaele Montella Energy analysis of a 4D Variational Data Assimilation algorithm and evaluation on ARM-based HPC systems - Rossella Arcucci, Davide Basciano, Alessandro Cilardo, Luisa D'Amore and Filippo Mantovani Track E: Special Session on Parallel Matrix Factorizations Chairpersons: Marian Vajtersic and G. Okša On the Tunability of a New Hessenberg Reduction Algorithm Using Parallel Cache Assignment - Mahmoud Eljammaly, Lars Karlsson and Bo Kågström New preconditioning for the one-sided block-Jacobi SVD algorithm - Gabriel Oksa, Martin Becka and Eva Vidlickova Structure-preserving of the block SS--Hankel method for solving Hermitian generalized eigenvalue problems - Akira Imakura, Yasunori Futamura and Tetsuya Sakurai On using the Cholesky QR method in the full-blocked one-sided Jacobi algorithm - Shuhei Kudo and Yusaku Yamamoto Track F: Main Track: Parallel Non-numerical Algorithms Chairperson: Przemyslaw Stpiczynski Two Parallelization Schemes for the Induction of Nondeterministic Finite Automata on PCs - Tomasz Jastrząb Approximating Personalized Katz Centrality in Dynamic Graphs - Eisha Nathan and David Bader Graph-Based Speculative Query Execution for RDBMS - Anna Sasak-Okoń and Marek Tudruj A GPU Implementation of Bulk Execution of the Dynamic Programming for the Optimal Polygon Triangulation - Kohei Yamashita, Yasuaki Ito and Koji Nakano
12:40 – 13:40	Lunch
13:40 – 15:40	Invited talks (3 slots x 40 min. in parallel) Track A: Chairperson: Jack Dongarra Sierra: The LLNL IBM CORAL System - Bronis R. de Supinski Enabling PDE software frameworks for exascale - Dominik Goeddeke Myths and Legends of Ancient Parallel Computing: Load Balancing - Alexey Lastovetsky Track B: Chairperson: Ewa Deelman Neuromorphic Computing: Architectures and Applications - Karheinz Meier Can we aggressively reduce scientific data without losing science? - Franck Cappello Challenges in Big Data Computing on HPC Platforms - Michela Taufer
15:40 – 16:10	Coffee break
16:10 – 17:25	Contributed papers in parallel (3 slots x 25 min.) Track A: WS on Power and Energy Aspects of Computations (PEAC) Chairperson: Masha Sosonkina GPU power modeling of HPC applications for the simulation of heterogeneous clouds - Antonios T. Makaratzis, Malik M. Khan, Konstantinos M. Giannoutakis, Anne C. Elster and Dimitrios Tzovaras Machine learning module for energy reduction in GPU-based supercomputers - Krzysztof Rojek and Roman Wyrzykowski Energy efficiency of ARMv8-based server chips in M2DC - a novel hyperscale microserver platform - Michał Kierzynka, Ariel Oleksiak, Chris Adeniyi-Jones and Javier Setoain Track B: WS on Language-Based Parallel Programming Models Chairperson: Ami Marowka Pardis: A Process Calculus for Parallel and Distributed Programming in Haskell - Christopher Blöcker and Ulrich Hoffmann Towards High-Performance Python - Ami Marowka Actor model of a new functional language - Anemone - Paweł Batko and Marcin Kuta Track C: WS on Complex Collective Systems Chairperson: Pawel Topa Application of Cellular Automata for Heuristic Optimization of Sensor Network Coverage Problem - Krzysztof Trojanowski, Artur Mikitiuk and Krzysztof J. M. Napiorkowski A Fuzzy Logic Inspired Cellular Automata Based Model for Simulating Crowd Evacuation Processes - Prodromos Gavriilidis, Ioannis Gerakakis, Ioakeim Georgoudas, Giuseppe Trunfio and Georgios Sirakoulis Nondeterministic Cellular Automaton for Modelling Urban Traffic with Self-organizing Control - Jacek Szklarski Track D: WS on Models, Algorithms and Methodologies for Hybrid Parallelism in new HPC Systems Chairpersons: Marco Lapegna, Giuliano Laccetti and Raffaele Montella Performance Assessment of the Incremental Strong Constraints 4DVAR Algorithm in ROMS - Rossella Arcucci, Luisa D'Amore, Yi Li, Andrey Moore, Luke Phillipson, Ralf Toumi and Raffaele Montella Evaluation of HCM: a new model to predict the execution time of regular parallel applications on a heterogeneous cluster - Thiago Marques Soares, Rodrigo Santos and Marcelo Lobosco Track E: Special Session on Parallel Matrix Factorizations Chairpersons: Marian Vajtersic and G. Okša Parallel Divide-and-Conquer Algorithm for Solving Tridiagonal Eigenvalue Problems on Manycore Systems - Yusuke Hirota and Toshiyuki Imamura Parallel Inverse of Non-Hermitian Block Tridiagonal Matrices - Louise Spellacy and Darach Golden Parallel Nonnegative Matrix Factorization Based on Newton Iteration with Improved Convergence Behavior - Rade Kutil, Markus Flatz and Marian Vajtersic
18:00	Travelling by buses to Zamoyski Palace in Kozłówka, excursion and barbecue

Tuesday, Sept. 12
8:30 – 9:50	Invited talks + industrial session (2 slots x 40 min in parallel) Track A: Chairperson: William Gropp Data Cyberinfrastructure for End-to-end Science: Experiences from the Ocean Observatories Initiative - Manish Parashar Challenges of Parallelizing Graph Algorithms for Network Science - Boleslaw Szymanski Track B (industrial session): Chairperson: Jeffrey Vetter Non-algorithmic approaches to improve application performance on Intel Xeon Phi processors - P. Uminski, J. Kogut, J. Zielinski, S. Szkoda (Intel) Lenovo servers based on Skylake CPUs - Even faster platform - Lenovo
9:50 – 10:20	Coffee break
10:20 – 12:25	Contributed papers in parallel (5 slots x 25 min.) Track A: WS on Power and Energy Aspects of Computations (PEAC) Chairperson: Krzysztof Kurowski Bi-cluster Parallel Computing in Bioinformatics - Performance and Eco-efficiency - Pawel Foszner and Przemyslaw Skurowski Applicability of the Empirical Mode Decomposition for Power Traces of Large-Scale Applications - Gary Lawson, Masha Sosonkina and Yuzhong Shen Performance and energy analysis of scientific workloads executing on heterogeneous processing elements - Anish Varghese and Alistair Rendell Energy efficient dynamic load balancing over multiGPU heterogeneous systems - Alberto Cabrera, Francisco Almeida, Alejandro Acosta and Vicente Blanco Track B: WS on Language-Based Parallel Programming Models Chairperson: Ami Marowka A Taxonomy of Task-Based Technologies for High-Performance Computing - Peter Thoman, Khalid Hasanov, Kiril Dichev, Roman Iakymchuk, Xavier Aguilar, Philipp Gschwandtner, Erwin Laure, Herbert Jordan, Pierre Lemarinier, Kostas Katrinis, Dimitrios Nikolopoulos and Thomas Fahringer Almost Optimal Column-wise Prefix-sum Computation on the GPU - Hiroki Tokura, Toru Fujita, Koji Nakano and Yasuaki Ito A Combination of Intra- and Inter-Place Work Stealing for the APGAS Library - Jonas Posner and Claudia Fohry Benchmarking Molecular Dynamics with OpenCL on Many-Core Architectures - Rene Halver, Godehard Sutmann and Wilhelm Homberg Efficient Language-Based Parallelization of Computational Problems Using Cilk Plus - Przemyslaw Stpiczynski Track C: Main Track: GPU Computing Chairpersons: Dominic Goeddeke and Robert Strzodka Experimental R-Trie for GPU - Krzysztof Kaczmarski and Albert Wolant A comparison of performance tuning process for different generations of NVIDIA GPUs and an example scientific computing algorithm - Krzysztof Banaś, Filip Krużel, Jan Bielański and Kazimierz Chłoń NVIDIA GPUs Scalability to Solve Multiple (Batch) Tridiagonal Systems. Implementation of cuThomasBatch - Pedro Valero-Lara, Ivan Martínez-Pérez, Raül Sirvent, Xavier Martorell and Antonio J. Peña Two-Echelon System Stochastic Optimization with R and CUDA - Witold Andrzejewski, Maciej Drozdowski, Gang Mu and Yongchao Sun Parallel Hierarchical Agglomerative Clustering for fMRI Data - Mélodie Angeletti, Jean-Marie Bonny, Franck Durif and Jonas Koko Track D: Main Track: Applications of Parallel Computing Chairperson: Raimondas Ciegis Global state monitoring in optimization of paralel, event-driven simulation - Lukasz Masko and Marek Tudruj High Performance Optimization of Independent Component Analysis Algorithm for EEG Data - Anna Gajos-Balińska, Grzegorz M. Wojcik and Przemyslaw Stpiczynski Continuous and discrete models of melanoma progression simulated in multi-GPU environment - Witold Dzwinel, Adrian Kłusek, Rafał Wcisło, Marta Panuszewska and Paweł Topa Early experience on using Knights Landing processors for Lattice Boltzmann applications - Enrico Calore, Alessandro Gabbana, Sebastiano Fabio Schifano and Raffaele Tripiccione Dynamic load balancing for solidification modeling on CPU-MIC hybrid platforms - Kamil Halbiniak, Lukasz Szustak, Adam Kulawik and Pawel Gepner Track E: MS on HPC Applications in Physical Sciences Chairpersons: Grzegorz Kamieniarz and Wojciech Florek Efficient Parallel Generation of Many-Nucleon Basis for Large-Scale Ab Initio Nuclear Structure Calculations - Daniel Langr, Tomáš Dytrych, Tomáš Oberhuber and František Knapp Parallel Exact Diagonalization Approach to Large Molecular Nanomagnets Modelling - Michał Antkowiak Cluster Monte Carlo method for the 3D Ashkin-Teller model - Zbigniew Wojtkowiak and Grzegorz Musial Application of numerical quantum transfer-matrix approach in the randomly diluted quantum spin chains - Ryszard Matysiak Track F: Main Track: Soft Computing with Applications Chairperson: Franciszek Seredynski Towards a Model of Semi-Supervised Learning for the Syntactic Pattern Recognition-based Electrical Load Prediction System - Janusz Jurek Parallel Processing of Color Digital Images for Linguistic Description of Their Content - Krzysztof Wiaderek, Danuta Rutkowska and Elisabeth Rakus-Andersson Co-evolution of fitness predictors and Deep Neural Networks - Paweł Koperek and Wlodzimierz Funika Performance evaluation of DBN learning on Intel multi- and manycore architectures - Tomasz Olas, Wojciech K. Mleczko, Robert Nowicki and Roman Wyrzykowski
12:25 – 14:00	Lunch
14:00 – 15:20	Invited talks (2 slots x 40 min. in parallel) Track A: Chairperson: Satoshi Matsuoka Tensor contractions and low-communication FFTs on GPUs - Cris Cecka Towards multi-objective, region-based auto-tuning for parallel programs - Thomas Fahringer Track B: Chairperson: Marek Tudruj Parallel Preemptive Online Scheduling with Deadlines and Slack - Uwe Schwiegelshohn Cloud Computing Uncertainty: opportunities and challenges - Andrei Tchernykh
15:20 – 15:40	Coffee break
15:40 – 18:10	Contributed papers in parallel (6 slots x 25 min.) Track A: Main Track: Numerical Algorithms and Parallel Scientific Computing Chairpersons: Gudula Rünger and Thomas Rauber Multilayer Approach for Joint Direct and Transposed Sparse Matrix Vector Multiplication for Multithreaded CPUs - Ivan Simecek, Ivan Kotenkov and Daniel Langr A Comparison of Soft-Fault Error Models in the Parallel Preconditioned Flexible GMRES - Evan Coleman, Aygul Jamal, Marc Baboulin, Amal Khabou and Masha Sosonkina Comparison of parallel time-periodic Navier--Stokes solvers - Peter Arbenz, Daniel Hupp and Dominik Obrist Blocked Algorithms for Robust Solution of Triangular Linear Systems - Carl Christian Kjelgaard Mikkelsen and Lars Karlsson A Comparison of Accuracy and Efficiency of Parallel Solvers for Fractional Power Diffusion Problems - Raimondas Ciegis Track B: Main Track: Environments and Frameworks for Parallel / Distributed / Cloud Computing Chairperson: Iosif Meerov Reproducible Computational Science - Challenges and Opportunities - Jarek Nabrzyski and Ian Taylor Dfuntest: A Testing Framework for Distributed Applications - Grzegorz Milka and Krzysztof Rzadca Security monitoring and analytics in the context of HPC processing model - Gerard Frankowski, Mikołaj Dobski, Norbert Meyer, Maciej Miłostan and Michał Pilc Multidimensional performance and scalability analysis for diverse applications based on system monitoring data - Dmitry Nikitenko, Pavel Shvets, Aleksey Teplov, Vadim Voevodin, Vladimir Voevodin, Nina Popova, Jonathan Bull, Ali Dorostkar, Sverker Holmgren, Anastasia Kruchinina and Maya Neytcheva Bridging the Gap between HPC and Cloud using HyperFlow and PaaSage - Dennis Hoppe, Yosandra Sandoval, Anthony Sulistio, Maciej Malawski, Bartosz Balis, Maciej Pawlik, Kamil Figiela, Dariusz Krol, Michal Orzechowski, Jacek Kitowski and Marian Bubak Track C: Main Track: Performance Evaluation of Parallel Algorithms and Applications Chairperson: Jan Kwiatkowski Early performance evaluation of the hybrid cluster with torus interconnect aimed at molecular-dynamics simulations - Vladimir Stegailov, Alexander Agarkov, Sergey Biryukov, Timur Ismagilov, Nikolay Kondratyuk, Evgeny Kushtanov, Dmitry Makagon, Anatoly Mukosey, Alexander Semenov, Alexey Simonov and Vyacheslav Vecher Load Balancing for CPU-GPU Coupling in Computational Fluid Dynamics - Immo Huismann, Matthias Lieber, Jörg Stiller and Jochen Fröhlich Implementation and Performance Analysis of 2.5D-PDGEMM on the K Computer - Daichi Mukunoki and Toshiyuki Imamura An Approach for Detecting Abnormal Parallel Applications Based on Time Series Analysis Methods - Denis Shaykhislamov and Vadim Voevodin Prediction of the Inter-Node Communication Costs of a New Gyrokinetic Code with Toroidal Domain - Andreas Jocksch, Noe Ohana, Emmanuel Lanti, Aaron Scheinberg, Stephan Brunner, Claudio Gheller and Laurent Villard D-Spline Performance Tuning Method Flexibly Responsive to Execution Time Perturbation - Guuing Fan, Akihiro Fujii, Teruo Tanaka and Takahiro Katagiri Track D: WS on Scheduling for Parallel Computing Chairperson: Maciej Drozdowski Scheduling data gathering with maximum lateness objective - Joanna Berlinska Fair Scheduling in Grid VOs with Anticipation Heuristic - Victor Toporkov, Dmitry Yemelyanov and Anna Toporkova A Security-Driven Approach to Online Job Scheduling in IaaS Cloud Computing Systems - Jakub Gasior, Franciszek Seredynski and Andrei Tchernykh Multi-objective Extremal Optimization in Processor Load Balancing for Distributed Programs - Eryk Laskowski, Ivanoe De Falco, Richard Olejnik, Umberto Scafuri, Ernesto Tarantino and Marek Tudruj Dynamic Load Balancing Algorithm for Heterogeneous Clusters - Tiago Marques Do Nascimento, Rodrigo Weber Dos Santos and Marcelo Lobosco Track E: WS on PGAS Programming Chairperson: Piotr Bala Interoperability of GASPI and MPI in the large scale Lattice-Boltzmann code - Roman Iakymchuk, Luis Cebamanos, Tiberiu Rotaru, Mirko Rahn, Erwin Laure, Valeria Bartsch, Christian Simmendinger and Stefano Markidis Evaluation of the parallel performance of the Java and PCJ on the Intel KNL based systems - Marek Nowicki, Łukasz Górski and Piotr Bala Fault-tolerance mechanisms for the Java parallel codes implemented with the PCJ library - Michał Szynkiewicz and Marek Nowicki Exploring graph analytics with the PCJ toolbox - Roxana Istrate, Panagiotis Barkoutsos, Michele Dolfi, Peter Staar and Costas Bekas Big Data analytics in Java with PCJ library - performance comparison with Hadoop - Marek Nowicki, Łukasz Górski, Magdalena Ryczkowska and Piotr Bala Performance comparison of graph BFS implemented in MapReduce and PGAS programming models - Magdalena Ryczkowska and Marek Nowicki Track F: MS on High Performance Computing Interval Methods Chairperson: Bartlomiej Kubica A new method for solving nonlinear interval and fuzzy equations - Ludmila Dymova and Pavel Sevastjanov Role of hull-consistency in the HIBA_USNE multithreaded solver for nonlinear systems - Bartlomiej Kubica Parallel computing of linear systems with linearly dependent intervals in MATLAB - Ondřej Král and Milan Hladík What Decision to Make In a Conflict Situation under Interval Uncertainty: Efficient Algorithms for the Hurwicz Approach - Bartlomiej Kubica, Andrzej Pownuk and Vladik Kreinovich Practical Need for Equation-Type Interval Equations and for Their Extended-Zero Solutions - Ludmila Dymova, Pawel Sewastjanow, Andrzej Pownuk and Vladik Kreinovich
18:40	Guided tour of Lublin Old Town and Conference Dinner in Royal Castle with short excursion to Chapel of the Holy Trinity

Wednesday, Sept. 13
8:40 – 10:00	Invited talks (2 slots x 40 min.) Chairperson: Bronis R. de Supinski Exploring Emerging Memory Technologies in Extreme Scale High Performance Computing - Jeffrey Vetter MPI+X for Extreme Scale Computing - William Gropp
10:00 – 10:20	Coffee break
10:20 – 12:00	Contributed papers in parallel (4 slots x 25 min.) Track A: Main Track: Numerical Algorithms and Parallel Scientific Computing Chairperson: Tomas Kozubek Efficient Cross Section Reconstruction on Modern Multi and Many Core Architectures - Yunsong Wang, François-Xavier Hugot, Emeric Brun, Fausto Malvagi and Christophe Calvin Parallel boundary element method for many-core architectures - Michal Kravcenko, Lukas Maly, Michal Merta and Jan Zapletal Stochastic bounds for Markov chains on Intel Xeon Phi coprocessor - Jaroslaw Bylina Strategy for Data-Flow Synchronizations in Stencil Parallel Computations on Intel KNL Processor - Lukasz Szustak, Kyrylo Khloponin and Roman Wyrzykowski Track B: Main Track: Environments and Frameworks for Parallel / Distributed / Cloud Computing Chairperson: Norbert Meyer A Memory Efficient Parallel All-Pairs Computation Framework: Using Computation - Communication Overlap - Venkata Kasi Viswanath Yeleswarapu and Arun Somani Automatic Parallelization of ANSI C to CUDA C programs - Jan Kwiatkowski and Dzanan Bajgorić Consistency Models for Global Scalable Data Access Services - Michał Wrzeszcz, Darin Nikolow, Tomasz Lichoń, Rafał Słota, Lukasz Dutka, Renata Slota and Jacek Kitowski Track C: Main Track: Task-Based Paradigm of Parallel Computing Chairperson: Rosa Badia TaskUniVerse: A Task-Based Unified Interface for Versatile Parallel Execution - Afshin Zafari Comparison of Time and Energy Oriented Scheduling for Task-based programs - Thomas Rauber and Gudula Ruenger Experiments with sparse Cholesky using a parametrized task graph implementation - Florent Lopez and Iain Duff A Task-Based Algorithm for Reordering the Eigenvalues of a Matrix in Real Schur Form - Mirko Myllykoski Track D: WS on Complex Collective Systems Chairperson: Renata Slota Towards Multi-Agent simulations accelerated using GPU - Kamil Piętak and Paweł Topa Tournament-based convection selection in evolutionary algorithms - Maciej Komosinski and Konrad Miazga Multi-Agent Systems programmed visually with Google Blockly - Szymon Górowski, Robert Maguda and Paweł Topa
12:00 – 13:20	Invited talks (2 slots x 40 min.) Chairperson: Jacek Kitowski Task-based parallel programming model in Python - Rosa Badia The curses and blessings of analytic performance modeling - Georg Hager
13:20 – 13:30	Closing remarks