logo conference

IXPUG Annual Fall Conference 2018

Conference Dates: September 25-28, 2018

Location: Intel Corporation (Jones Farm Conference Center, JFCC Building, 2111 NE 25th Avenue Hillsboro, Oregon 97124). Please reference the the Travel Document for transportation details (campus map, parking, etc.) and other fun facts about Oregon and Intel Corporation's campus.

Cost: This is a free and open to the public training conference. The event is sponsored by Intel and all compliance certification applies: Intel is committed to compliance with all applicable laws, regulations, policies and ethics rules. This notice is intended to ensure that your participation in the IXPUG Annual Fall Conference 2018 held at Intel will not violate any such laws, regulations, policies or rules. Intel plans to serve a continental breakfast, modest lunch, and afternoon refreshment on each day of the event. Representatives of government agencies or government-funded entities such as national laboratories can pay for their meals prior to the event HERE (please follow these instructions) or onsite by credit card. Intel seeks no promises or favoritism for itself or any of its affiliates in exchange for the hospitality provided.

Event Description: This IXPUG conference is focused on all aspects of employing and adopting many-core processing technologies and techniques for optimal application execution, including topics that cover system hardware beyond the processor (memory, interconnect, etc.), software tools, programming models, new workloads (visualization, data analytics, machine learning, etc.) and more. The conference will provide an interactive experience focused on key topics associated with high-performance computing, artificial intelligence, data analytics, cloud computing, and more. The meeting is being co-organized by Sandia National Laboratory and Los Alamos National Laboratory, and will be held at the Intel Corporation (Hillsboro, Oregon) facility. The conference activities include a keynote, plenary sessions, 30-minute technical lectures, 10-15 minute Lightning Talks, Posters and Tutorials. Attendees will experience an open forum, where Intel and industry experts will share best practices and techniques used for maximizing software efficiency using Intel architecture with fellow application programmers, software developers, scientist, researchers, academic, students, data analyst, etc. The challenges surrounding application performance and scalability will be covered across all levels, including application tuning on large HPC systems.

Event Agenda: Presentations and video archive links will be posted as soon as possible.

Day 1: Tuesday, September 25, 2018

Start	End	Title	Author(s)	Presentation	Video
08:30	09:00am	Registration & Coffee/Tea
09:00	09:15am	Welcome by Host	Clay Hughes and David Martin	Presentation	Video
09:15	09:45am	Site Update	Dan Stanzione (TACC)	Presentation	Video
		Lightning Talks 1: Optimization
9:45	10:45am	- Intel® Omni-Path Architecture Multiple Endpoints -Intel(R) Omni-Path Architecture Performance Optimization -Exploring a nontrivial upper bound for pipelined double-precision FMA instructions on Knight’s Landing -Understanding System Performance using Intel® VTune™ Amplifier’s Platform Profiler	-Edward Mascarenhas, James Erwin -Edward Mascarenhas, James Erwin & Kevin Pine -Damon McDougall & John McCalpin -Munara Tolubaeva & Bhanu Shankar	LT 1:1 Presentation LT 1:2 Presentation LT 1:3 Presentation LT 1:4 Presentation	LT 1:1 Video LT 1:2 Video LT 1:3 Video LT 1:4 Video
10:45	11:00am	Break
11:00	12:00pm	Keynote: "Scaling Artificial Intelligence from HPC to Cloud"	Pradeep Dubey, Intel	Presentation	Video
12:00	13:00pm	Lunch
		Technical Session 1: Tools (Session Chaired by Thomas Steinke)
13:00	13:30pm	Flipping Tables: Favoring computation over memoization for performance	Jason Sewall	TS 1:1 Presentation	TS 1:1 Video
13:30	14:00pm	Analyzing OpenMP and TBB task graphs using Intel® Advisor – Flow Graph Analyzer (FGA)	Vishakha Agrawal, Mark Lubin, Pablo Reble, Vasanth Tovinkere & Mike Voss	TS 1:2 Presentation	TS 1:2 Video
14:00	14:30pm	Optimizing Iso3DFD with Intel® Advisor and Intel® VTune™ Amplifier	Carlos Rosales-Fernandez & Kevin Oleary	TS 1:3 Presentation	TS 1:3 Video
14:30	15:00pm	Configuration-Free Profiling at Scale	Carlos Rosales-Fernandez, Bei Wang, & Dmitry Prohorov	TS 1:4 Presentation	TS 1:4 Video
15:00	15:15pm	Break
		Technical Session 2: Optimization (Session Chaired by John Pennycook)
15:15	15:45pm	Designing High Performance MPI Libraries for the Multi-/Many-core Era	Dhabaleswar Panda, Jahanzeb Maqbool Hashmi, Sourav Chakraborty, Mohammadreza Bayatpour & Hari Subramoni	TS 2:1 Presentation	TS 2:1 Video
15:45	16:15pm	Address Hashing in Intel KNL and SKX Processors	John McCalpin	TS 2:2 Presentation	TS 2:2 Video
16:15	16:45pm	Optimization of TensorFlow Serving Application on Intel Xeon Scalable Processors	Vamsi Sripathi and Vikram Saletore	TS 2:3 Presentation	TS 2:3 Video
16:45	17:00pm	IXPUG Day 1 Closing Remarks (Day 1 Recap & Day 2 Agenda)	David Gunter and Clay Hughes
17:00	18:30pm	Networking Reception & Refreshments

Day 2: Wednesday, September 26, 2018

Start	End	Title	Author(s)	Presentation	Video
08:30	09:00am	Registration & Coffee/Tea
09:00	09:15am	Day 2 Agenda Review	David Gunter and Clay Hughes
09:15	10:15am	Keynote: "SPARTA Heterogeneous Full Trinity Runs: Successes and Challenges"	Stan Moore, Sandia National Laboratories	Presentation	Video
10:15	10:30am	Break
		Technical Session 3: Application Experience (Session Chaired by Jeff Hammond)
10:30	11:00am	Using MPI+Kokkos to Portably Extend Uintah to Many Core Systems	John Holmen	TS 3:1 Presentation	TS 3:1 Video
11:00	11:30am	DM-HEOM: A Case Study in Performance Portability	Matthias Noack	TS 3:2 Presentation	TS 3:2 Video
11:30	12:00pm	Optimization of the Model for Prediction Across Scales: Ocean Core targeting Production Scale Use of Knights Landing Processor Architecture	William Arndt	TS 3:3 Presentation	TS 3:3 Video
12:00	13:00pm	Lunch
		Lightning Talks 2: Applications
13:00	14:00pm	-Enabling ocean modeling code MPAS-Ocean on Intel® Xeon® -Improved threading performance of Quantum ESPRESSO -Performance Optimization Techniques for Accelerating WRF Physics Codes on Intel Micro-Architectures -Application Study of Gyrokinetic PIC codes on Intel KNL architecture	-Nitya Hariharan, Jason Do, Mark Roger Petersen, and Phil Jones -Ye Luo -Timbwaoga Ouermi, Robert M. Kirby, and Martin Berzins -Brian Mackie-Mason, Paulius Velesko, Robert Hager, C-S Chang, and Timothy J. Williams	LT 2:1 Presentation LT 2:2 Presentation LT 2:3 Presentation LT 2:4 Presentation	LT 2:1 Video LT 2:2 Video LT 2:3 Video LT 2:4 Video
		Technical Session 4: Machine Learning (Session Chaired by Richard Gerber and Melyssa Fratkin)
14:00	14:30pm	Distributed Training of 3D Generative Adversarial Networks for Fast Detector Simulation on Intel® Xeon® HPC Cluster	Vikram Saletore, Sofia Vallecorsa, Damian Podareanu, Federico Carminati, Valeriu Codreanu, Gulrukh Khattak, and Hans Pabst	TS 4:1 Presentation	TS 4:1 Video
14:30	15:00pm	Performance Optimization of Multi-scale CNN Training with High Content Cellular Imaging on Intel® Xeon® Scalable Processors	Kushal Datta, Sun Choi, Vikram Saletore, and Kyle Ambert	TS 4:2 Presentation	TS 4:2 Video
15:00	15:15pm	Break
15:15	15:45pm	A Machine Learning Framework for Large-Scale Weather and Climate Prediction using Exact and Approximate Linear Algebra Computation	Sameh Abdulah, Hatem Ltaief, Ying Sun, Marc Genton, and David Keyes	TS 4:3 Presentation	TS 4:3 Video
15:45	16:15pm	David Fast & Accurate Training of an AI Radiologist with Deep Learning on Intel Xeon-based DellEMC Supercomputers	Vikram Saletore, Valeriu Codreanu & Damian Podareanu	TS 4:4 Presentation	TS 4:4 Video
16:15	16:45pm	Site Updates	Richard Gerber (NERSC/LBNL), Thomas Steinke (Zuse Institute Berlin) & David Martin (Argonne National Laboratory)	Presentations: NERSC ZIB ALCF	Video
16:45	17:00pm	IXPUG Day 2 Closing Remarks (Day 2 Recap & Day 3 Agenda)	David Gunter and Clay Hughes

Day 3: Thursday, September 27, 2018

Start	End	Title	Author(s)	Presentation	Video
08:30	09:00am	Registration & Coffee/Tea
09:00	09:15am	Day 3 Agenda Review	David Gunter and Clay Hughes
		Lightning Talks 3:
09:15	10:15am	-Improving Oil and Gas Extraction Simulation Performance using Intel® Xeon® and Xeon Phi™ Architectures -Project DisCo: Physics-based discovery of coherent structures in spatiotemporal systems -Optimization strategy for MASNUM surface wave model -OpenMP for Reconfigurable Heterogeneous Architectures	-Matheus Serpa, Eduardo Cruz, and Philippe Navaux -Adam Rupe, Nalini Kumar, Karthik Kashinath, James P. Crutchfield, Ryan G. James, Mr. Prabhat, and Victor Lee -Zhenya Song -Marius Knaust and Florian Mayer	LT 2:1 Presentation LT 2:2 Presentation LT 2:3 Presentation LT 2:4 Presentation	LT 1:1 Video LT 1:2 Video LT 1:3 Video LT 1:4 Video
10:15	10:30am	Break
		Technical Session 5: Vectorization (Session Chaired by David Martin)
10:30	11:00am	Intel Vector Math Libraries – Features, Usage, and Performance Improvements	Marius Cornea, Cristina Anderson, Robin Gilbert, Ahmet Akkas, and Andrey Kolesov	TS 5:1 Presentation	TS 5:1 Video
11:00	11:30am	SIMD-vectorization in VASP	Florian Wende	TS 5:2 Presentation	TS 5:2 Video
11:30	12:00pm	OpenMP SIMD support for both Inclusive and Exclusive Scans	Anoop Madhusoodhanan Prabha and Nikolay Panchenko	TS 5:3 Presentation	TS 5:3 Video
12:00	12:15pm	IXPUG Day 3 Closing Remarks
12:15	13:00pm	Lunch
		Tutorials (2 Tracks)
13:00	15:00pm	Tutorial Track 1: Performance Profiling in the Cloud	Jennifer DiMatteo	TT 1:1 Presentation	No Video
13:00	15:00pm	Tutorial Track 2: Forwarding Mindset	Csaba Keszei	TT 2:1 Presentation	No Video
15:00	15:15pm	Break
		Tutorials (2 Tracks)
15:15	17:15pm	Tutorial Track 1: How to rapidly develop high-performance stencil code via the YASK framework (a tutorial)	Charles Yount and Joonmoo Huh	TT 1:2 Presentation	No Video
15:15	17:15pm	Tutorial Track 2: Verificarlo: Floating point computing validation on new architecture and large scale systems	Eric Petit, Pablo de Oliveira Castro, and Yohan Chatelain	TT 2:2 Presentation	No Video
17:15	17:20pm	IXPUG Day 3 Closing Remarks (Day 3 Recap & Day 4 Agenda)	Lisa Smith

Day 4: Friday, September 28, 2018

Start	End	Title	Author(s)	Presentation	Video
08:30	09:00am	Registration & Coffee/Tea
		Tutorials (2 Tracks)
09:00	11:00am	Tutorial Track 1: Intel Omni-Path Architecture Fabric Introduction	Michael Savasta and Lester Dimit	TT 1:3 Presentation	No Video
09:00	11:00am	Tutorial Track 2: Optimizing for Multi/Manycore (SKX/KNL): Multithreading with OpenMP	W Cyrus Proctor and Antia Lamas-Linares	TT 2:3 Presentation	No Video
11:00	11:05	IXPUG Day 4 Closing Remarks (Day 4 Recap & Upcoming Activities)	Lisa Smith

Travel Information:

Visa Requirements: Please contact This email address is being protected from spambots. You need JavaScript enabled to view it. for visa inquiries.
Transportation and Parking: Check out the Travel Document for transportation details (nearby airport, etc.), fun facts about Oregon and Intel Corporation's campus.
Lodging: Nearby hotel suggestions are listed below.


Hotel	Address
Larkspur Landing Hillsboro	3133 NE Shute Rd, Hillsboro, OR 97124; (503) 681-2121
Residence Inn by Marriott Portland Hillsboro/Brookwood	3160 NE Brookwood Pkwy, Hillsboro, OR 97124; (503) 844-7777
Holiday Inn Express Portland West/Hillsboro	5900 NE Ray Cir, Hillsboro, OR 97124; (503) 844-9696
Comfort Inn	3500 NE Cornell Rd, Hillsboro, OR 97124; (503) 648-3500
TownePlace Suites by Marriott Portland Hillsboro	6550 NE Brighton St, Hillsboro, OR 97124; (503) 268-6000
The Orenco, Boutique Hotel	1457 NE Orenco Station Pkwy, Hillsboro, OR 97124; (888) 503-7094
Embassy Suites by Hilton Portland Hillsboro, Oregon	20001 NW Tanasbourne Dr, Hillsboro, OR 97124; (503) 718-0007
Courtyard by Marriott Portland Hillsboro	3050 NW Stucki Pl, Hillsboro, OR 97124; (503) 690-1800
SpringHill Suites by Marriott Portland Hillsboro	7351 NE Butler St, Hillsboro, OR 97124; (503) 547-0202
Holiday Inn Hillsboro	2575 NW Aloclek Dr, Hillsboro, OR 97124; (503) 640-1745
Residence Inn by Marriott Portland Hillsboro	18855 NW Tanasbourne Dr, Hillsboro, OR 97124; (503) 531-3200
Extended Stay America - Portland - Hillsboro	19311 NW Cornell Rd, Hillsboro, OR 97124; (503) 439-0706
Extended Stay America - Portland - Beaverton/Hillsboro - Eider Ct	18665 NW Eider Ct, Beaverton, OR 97006; (503) 439-1515
Aloft Hillsboro-Beaverton	1705 NW Amberglen Ct, Hillsboro, OR 97006; (503) 277-1900

Call for Abstracts: IXPUG is soliciting submissions for technique presentations on innovative work using Intel architecture from users in academia, industry, government/national labs, etc. describing original discoveries, experiences and methods for obtaining efficient and scalable use of heterogeneous systems. IXPUG welcomes technical submissions that cover 30 minute lectures, 10-15 minute lightning talks and/or 120 minute tutorials. Please notice, tutorials should be interactive focusing on key topics of high-performance computing, interconnects, artificial intelligence, cloud or performance evaluation.

Abstract Submission Guidelines: A short abstract should be submitted by July 13, 2018 and the content should reflect the topics of interest that is listed below. Please use the Abstract Template, as you convey your content and submit your abstract to IXPUG EasyChair. All final presentations are due by September 24, 2018. 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: Algorithms & Methods, Compiler Flags, Software Environment & Tools, Libraries, Parallel- Programming (Communications, Thread & Process Management Experience, All), Multi- node, Memory, Vectorization
Products: Intel® Xeon Phi™ Co/Processor, Intel® Xeon Processor Family, Intel® Omni Path Fabric, Intel® FPGA, Intel® SSDs/NVMe Solutions, Intel® Lustre Software, Visualization Technology, Intel® SW Tools
Domains: Artificial Intelligence (Machine Learning/Deep Learning), Astrophysics, Bioinformatics, Chemistry, Climate & Weather, Computational Fluid Dynamics, Data Analytics, Energy/Oil & Gas, Financial Services, Geophysics, Libraries & Tools, Life Sciences, Material Science, Medical imaging, Molecular Dynamics, Nanotechnology, Physics, Visualization, etc.

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 Languagle Development)
Experience using extreme scale systems (Usability, In-situ Visualization, Programming Challenges, Algorithms and Methods)
Application characterization on emergining technologies (Novel Memories (HBM and NVM)), processors (Xeon, Xeon Phi, Xeon+FPGA, Cloud Computing, etc.)

Important Dates:

Abstract Submission Timelines	March 14, 2018 - July 13, 2018
Final Date to Submit	July 13, 2018
Acceptance Notification	July 27, 2018
Agenda Posted	August 6, 2018
Registration Deadline	September 14, 2018
Speaker's Final Presentations Uploaded to EasyChair	September 24, 2018

Event Planning Team:

Name	Institute
Clay Hughes	Sandia National Laboratory
Hai Ah Nam	Los Alamos National Laboratory
John Pennycook	Intel Corporation
Lisa M. Smith	Intel Corporation
Melyssa Fratkin	Texas Advanced Computing Center

Abstract Review Committee:

Name	Institute
Clay Hughes	Sandia National Laboratory
David Keyes	King Abdullah University of Science & Technology
David Martin	Argonne National Laboratory
Estela Suarez	Forschungszentrum Juelich / Juelich Supercomputing Centre
Fabio Affinito	CINECA
Hai Ah Nam	Los Alamos National Laboratory
James Lin	Shanghai Jiao Tong University
John Pennycook	Intel Corporation
Kent Milfeld	Texas Advanced Computing Center
Melyssa Fratkin	Texas Advanced Computing Center
Richard Gerber	National Energy Research Scientific Computing Center, Lawrence Berkeley National Lab
Taisuke Boku	University of Tsukuba, Japan
Thomas Steinke	Zuse Institute Berlin, Germany
Vit Vondrak	VSB - Technical University of Ostrava, Czech Republic