IXPUG Annual Conference 2019

Conference Dates: September 24-27, 2019

Location: Globe of Science and Innovation at CERN (Geneva, Switzerland)

Registration: Register now! Seating is limited. Advance registration is required.

 

Event Description:

This IXPUG conference is focused on all aspects of adopting and employing state-of-the-art data-processing technologies and techniques for optimal application execution. This includes accelerators (e.g. FPGAs, GPUs), as well as topics related to system hardware beyond the processor (memory, interconnect, etc.), software tools, programming models, new workloads, and more — all with a focus on Intel platforms. The conference will provide an interactive experience, organised around key themes associated with high-performance computing, data analytics, artificial intelligence (machine and deep learning), cloud computing, and more. The meeting will be held at CERN, the European Laboratory for Particle Physics, in Geneva, Switzerland. The conference will feature keynote presentations, invited talks, 30-minute technical sessions, 10-to-15-minute lightning talks, and two-hour hands-on tutorials. Attendees will encounter an open forum, through which experts from academia and industry including Intel representatives will share best practices and techniques for maximizing software efficiency. We will share experience in using Intel architectures and technologies with fellow application programmers, software developers, scientists, researchers, academics, students, systems analysts, etc. The challenges surrounding application performance and scalability will be covered across all levels, including application tuning on large HPC systems.

Event Agenda:

Session dates and times are subject to change. Presentations and recording archive links will be posted as soon as possible following the sessions. 

Day 1: Tuesday, September 24, 2019

Start End Title Author(s) Presentation Recording
 08:30 09:00 Registration and Coffee      
09:00 09:15 Welcome by Host Alberto Di Meglio (CERN), Eckhard Elsen (CERN), Thomas Steinke (IXPUG President; Zuse Institute Berlin)    
09:15 10:00 Deep Learning Workflows: Examples from High Energy Physics Sofia Vallecorsa (CERN)    
10:00 10:30 Deploying AI Frameworks on Secure HPC Systems with Containers Walter Riviera (Intel Corporation)    
10:30 11:00 Distributed Training of Generative Adversarial Networks for Fast Simulation

Sofia Vallecorsa (CERN)

   
11:00 11:15 Performance and Scalability Analysis of CNN-based Deep Learning Inference in the Intel Distribution of OpenVINO Toolkit Iosif Meyerov (Lobachevsky State University of Nizhni Novgorod), Valentina Kustikova (Lobachevsky State University of Nizhni Novgorod), Evgenii Vasiliev (Lobachevsky State University of Nizhni Novgorod), Evgeniy Kozinov (Lobachevsky State University of Nizhni Novgorod), Valentin Volokitin (Lobachevsky State University of Nizhni Novgorod), Zakhar Matveev (Intel Corporation), Kirill Korniakov (Intel Corporation)    
11:15 11:30 Coffee Break      
11:30 12:15

Progress and Challenges on HPC and AI Convergence

Victor Lee (Intel Corporation)    
12:15 13:15 Lunch Break      
13:15 13:45 “Big Data In HEP” - Physics Data Analysis, Machine learning and Data Reduction at Scale with Apache Spark Luca Canali (CERN), Vaggelis Motesnitsalis (CERN), Oliver Gutsche (Fermilab)    
13:45 14:15 I/O for Deep Learning at Scale Quincey Koziol (National Energy Research Scientific Computing Center, Lawrence Berkeley National Lab)    
14:15 14:30 Deep Learning for Satellite Imagery Gul Rukh Khattak (CERN), Sofia Vallecorsa (CERN), Taghi Aliyev (CERN)    
14:30 15:00 Exploiting Emerging Multi-core Processors for HPC and Deep Learning using MVAPICH2 MPI Library Dhabaleswar Panda (The Ohio State University), Hari Subramoni (The Ohio State University)    
15:00 15:30

OpenCL in Scientific High Performance Computing

Matthias Noack (Zuse Institute Berlin)    
15:30 15:45 Coffee Break      
15:45 16:15

OpenMP to FPGA Offloading Prototype Using the Intel FPGA SDK for OpenCL – an IXPUG Success Story

Marius Knaust (Zuse Institute Berlin)    
16:15 17:00 Next Generation Intel MPI Product for Next Generation Systems: Latest Intel MPI Features and Optimization Techniques Dmitry Durnov (Intel Corporation)    
17:00 17:15

IXPUG Day 1 Closing Remarks (Day 1 Recap & Day 2 Agenda)

 

   
17:15 19:00  Networking Reception      

Day 2: Wednesday, September 25, 2019

Start End Title Author(s) Presentation Recording
 08:30 09:00

Registration and Coffee

     
09:00 09:15

Day 2 Agenda Review

     
09:15 09:45

Cost-Efficiency of Large-Scale Electronic Structure Simulations with Intel Xeon Phi Processors

Hoon Ryu (Korea Institute of Science and Technology Information)    
09:45 10:15 Optimizing Beyond Vectorization and Parallelization: a Case Study on QMCPACK William Jalby (UVSQ / ECR), Cédric Valensi (UVSQ / ECR)    
10:15 10:30 Hybrid-BLonD: Efficient Scale-out of Beam Longitudinal Dynamics Simulations Konstantinos Iliakis (CERN & NTUA/MicroLab)    
10:30 10:45 Optimizing Astrophysical Simulation and Data Analysis Codes on Intel Architectures

Salvatore Cielo (Leibniz Supercomputing Centre), Luigi Iapichino (Leibniz Supercomputing Centre), Fabio Baruffa (Intel Corporation)

   
10:45 11:00 Coffee Break      
11:00 11:15

High-Fidelity Rendering for Large-Scale Tiled Displays

Joao Barbosa (The University of Texas at Austin - TACC), Paul Navratil (The University of Texas at Austin- TACC)    
11:15 11:30

Applying Vectorization to Lattice QCD Calculations

Shun Xu (Computer Network Information Center, Chinese Academy of Sciences), Zhong Jin (Computer Network Information Center, Chinese Academy of Sciences)

   
11:30 12:15

A Journey over the Memory Management Stack for HPC Large Applications on Modern Architectures

Sébastien Valat (ATOS/Bull)    
12:15 13:15 Lunch Break      
13:15 14:15 Site Updates

Joao Barbosa (Texas Advanced Computing Center)


Estela Suarez (Juelich Supercomputing Centre)


Salvatore Cielo (Leibniz Supercomputing Centre)


Thomas Steinke (Zuse Institute Berlin)

   
14:15 14:45

Disrupting High Performance Storage with Intel® Optane™ DC Persistent Memory & DAOS

John Carrier (Intel Corporation)    
14:45 15:15

Modernizing Legacy Codes for Next-Generation Storage Infrastructures: A Case Study of PALM on Intel DAOS

Steffen Christgau (Zuse Institute Berlin)

   
15:15 15:30

Toward Easing the Use of Optane DIMMs as Part of Heterogeneous Memory Systems

Marc Jordà (Barcelona Supercomputing Center), Harald Servat (Intel Corporation), Antonio J. Peña (Barcelona Supercomputing Center)

   
15:30 15:45

Coffee Break

     
15:45 16:15

Persistent Memory Based Key-Value Store for Data Acquisition Systems

Maciej Maciejewski (Intel Corporation), Grzegorz Jereczek (Intel Corporation), Danilo Cicalese (CERN), Giovanna Lehmann Miotto (CERN)

   
16:15 16:45

Partitioned Interleaved Bloom Filters Using Optane DC Persistent Memory

Knut Reinert (Freie Universität Berlin), Enrico Seiler (Freie Universität Berlin)    
16:45 17:30

Exploiting Persistent Memory for Workflows and Computational Simulation

Adrian Jackson (The University of Edinburgh)    
17:30 17:45 IXPUG Day 2 Closing Remarks (Day 2 Recap and Day 3 Agenda)      

Day 3: Thursday, September 26, 2019

Start End Title Author(s) Presentation Recording
 08:30 09:00 Registration and Coffee      
09:00 09:15 Day 3 Agenda Review      
09:15 09:45

An Integrated FPGA Orchestrator for Seamless Scalability and Resource Management

Elias Koromilas (InAccel)    
09:45 10:15

Characterizing Performance Benefits of HBM2 on Intel® Stratix® 10 FPGAs

Richard Chamberlain (BittWare), Tiziano De Matteis (ETH Zurich)    
10:15 10:45

Acceleration of Scientific Deep Learning Models on Heterogeneous Computing Platform with Intel® FPGA

Chao Jiang (University of Florida), Sofia Vallecorsa (CERN)

   
10:45 11:00

Coffee Break

     
11:00 11:30

Natural Language Processing with Intel® Quantum Simulator

Venkatesh Kannan (Irish Centre for High-End Computing (ICHEC))

   
11:30 11:45

Simple Implementation of Quantum Bits in Silicon by Decoupling Them in Space and Time

Nagi Mekhiel (Ryerson University)

   
11:45 12:30

Extraordinary Use Cases for Exascale Class Technology: How AI and DA Combined with HPC Shape Innovation in Exascale Architectures

Marie-Christine Sawley (Intel Corporation)

   
12:30 13:15

General Purpose Heterogenous Next-Generation Systems

Richard Gerber (National Energy Research Scientific Computing Center, Lawrence Berkeley National Lab), Nicholas Wright (National Energy Research Scientific Computing Center, Lawrence Berkeley National Lab)     
 13:15 13:30 IXPUG Day 3 Closing Remarks (Day 3 Recap and Conference Closing Remarks)       
 13:30 14:45  Lunch Break      
 14:45 16:45

Tutorial 1: Numba/HPAT and Daal4py: The Painless Route in Python to Fast and Scalable Data-Analytics/Machine-Learning

 

Frank Schlimbach (Intel Corporation)

   
 14:45 16:45 Tutorial 2: Verificarlo: Floating-Point Computing Verification and Optimization Eric Petit (Intel Corporation), Yohan Chatelain (Universty of Versailles, Li-Parad, ECR Lab), Pablo de Oliveira Castro (Universty of Versailles, Li-Parad, ECR Lab), David Defour (Universty of Versailles, ECR Lab, and University of Perpignan)     
17:00 18:00 Visit to CERN Supercyclotron Limited attendance; details to sign up provided separately to conference registrants    

Day 4: Friday, September 27, 2019

Start End Title Author(s) Presentation Recording
 09:00 12:00 Tutorial 3: Compute Offload Acceleration with FPGA  Francisco Perez (Intel Corporation)    
12:00 13:30

Lunch Break

     
13:30 16:00 Tutorial 4: Intel Solution for AI Walter Riviera (Intel Corporation)    

Travel Information:

  • CERN Tours: Explore CERN with guided tours for individuals and permanent exhibitions.
  • Visa Requirements:  Please contact This email address is being protected from spambots. You need JavaScript enabled to view it. at CERN openlab for visa questions.
  • Transportation and Parking: Please review the "Visit CERN" website for transportation and parking details (including nearby airports, public transportation, CERN site map, etc.). Also, review the "Code of Conduct for Visitors to CERN" and "Frequently Asked Questions" for more details.

  • Lodging: Nearby hotel suggestions are listed below

Hotels in Geneva, within walking distance to Tram 18 to CERN:

NH Hotel Geneva Airport 

Holiday Inn Express Geneva Airport

Ibis Budget Hotel Geneva Airport

Ibis Styles Geneve Mont Blanc

Hotels on the French side (these hotels have shuttles from the airport, and in the morning they can take guests and drop them off at CERN):

Business Park Hotel Thoiry

Hotel Ibis St Genis Pouilly

Call for Abstracts

IXPUG is soliciting submissions for technical presentations on innovative work from users in academia, industry, government/national labs, OEMs, ISVs, etc. describing original discoveries, experiences and methods for obtaining efficient and scalable use of heterogeneous systems using Intel architectures and technology platforms. IXPUG welcomes technical abstracts that cover 30-minute technical sessions, 10-to-15 minute lightning talks, and two-hour hands-on tutorials. Please notice, tutorials should be interactive focusing on key topics of high-performance computing, data analytics, artificial intelligence (machine and deep learning, visualization, cloud or performance evaluation, etc.).

Abstract Submission Guidelines:

A short abstract should be submitted by July 19, 2019 (updated!) via EasyChair and the content should reflect the topics of interest that are listed below. All final presentations are due by September 23, 2019. We would like you to include keywords that pertain to the techniques, Intel products and associated domains that pertain to your technical work. Please see the full list below:

  • Techniques:  Artificial Intelligence (Machine Learning/Deep Learning), Algorithms & Methods, Compiler Flags, Software Environment & Tools, Libraries & Tools, Parallel- Programming (Communications, Thread & Process Management Experience, All), Multi-node, Memory, Vectorization, etc.

  • Products: Intel® Xeon Scalable processor, Intel® Omni Path Fabric, Intel® FPGA, Intel® SSDs/NVMe Solutions, Intel® Lustre Software and DAOS, Visualization Technology, Intel® SW Tools 

  • Domains: Astrophysics, Bioinformatics, Chemistry, Climate & Weather, Computational Fluid Dynamics, Data Analytics, Energy/Oil & Gas, Financial Services, Geophysics, Life Sciences, Material Science, Medical imaging, Molecular Dynamics, Nanotechnology, Physics, Visualization, High Energy Physics, etc.

Abstract Topics of Interest:

  • Implication of workload behavior on system design at extreme scale (Power, Reliability, Scalability, Performance, Processor Design, Memory System, I/O)

  • Software environments and tools for computing at extreme scale (Instrumentation, Debugging/Correctness, Thread and Process Management, Libraries and Language Development)

  • Experience using extreme scale systems: Usability, In-situ Visualization, Programming Challenges, Algorithms and Methods, etc.

  • Application characterization on emerging technologies: Novel Memories (NVM), processors (Intel® Xeon Scalable processor, Intel® FPGA, etc.) 

Organizers

  • Thomas Steinke, IXPUG President (Zuse Institute Berlin)
  • Estela Suarez (Juelich Supercomputing Centre)
  • R. Glenn Brook, IXPUG Vice-President (University of Tennessee, Knoxville)
  • Alberto Di Meglio (CERN openlab)
  • Maria Girone (CERN openlab)
  • Federico Carminati (CERN openlab)
  • Sofia Vallecorsa (CERN openlab)
  • Andrew Robert Purcell (CERN openlab)
  • Kristina Gunne (CERN openlab)
  • David Martin (Argonne National Laboratory)
  • Richard Gerber (NERSC/LBNL)
  • Nalini Kumar (Intel Corporation)
  • Shari Lawrence (Intel Corporation)

Abstract Review Committee

  • Thomas Steinke, IXPUG President (Zuse Institute Berlin)
  • Estela Suarez (Juelich Supercomputing Centre)
  • R. Glenn Brook, IXPUG Vice-President (University of Tennessee, Knoxville)
  • Alberto Di Meglio (CERN openlab)
  • Maria Girone (CERN openlab)
  • Federico Carminati (CERN openlab)
  • Sofia Vallecorsa (CERN openlab)
  • Nalini Kumar (Intel Corporation)
  • David Martin (Argonne National Laboratory)
  • Richard Gerber (NERSC/LBNL)
  • Hai Ah Nam (LANL)
  • Clay Hughes (SNL)
  • Fabio Affinito (CINECA)
  • John Pennycook (Intel Corporation)
  • David Keyes (KAUST)
  • Taisuke Boku (University of Tsukuba)
  • James Lin (Shanghai Jiao Tong University)
  • Sergi Siso (UK Science & Technology Facilities Council)

Sponsors:

  • CERN openlab
  • Intel Corporation

Questions? Email This email address is being protected from spambots. You need JavaScript enabled to view it.