2019 ALCF Simulation, Data, and Learning Workshop

2019 ALCF Simulation, Data, and Learning Workshop

 

 

 

 

 

 

 

 

From October 1–3, 2019, we will host the ALCF Simulation, Data, and Learning Workshop to help researchers improve the performance and productivity of simulation, data science, and machine learning applications on ALCF systems. Workshop participants will have the opportunity to:

  • Work directly with ALCF and industry professionals during dedicated hands-on sessions
  • Learn how to use available tools and frameworks to improve productivity
  • Learn how to set up workflows and use containers
  • Test and debug codes with exclusive full-system reservations on ALCF systems
  • Get assistance with Director's Discretionary projects to help prepare for a major allocation award
  • Improve the performance of existing ALCF projects
  • Plan ahead for 2020-2021 allocation proposal submissions

REGISTRATION DEADLINES

  • Foreign Nationals: Friday, September 6, 2019
  • U.S. Citizens: Friday, September 13, 2019

Note: Registrants will be reviewed for experience level and will be asked to provide goals for attending.

 

 

Agenda

Day 1 (Tuesday, October 1)

8:30 Shuttle Departs from Guest House to TCS, Bldg. 240
9:00-9:10 Welcome - Michael Papka, ANL
9:10-9:15 Introduction and Safety Review - Yasaman Ghadar, ANL
9:15-10:00 Getting Started with ALCF Systems - Adrian Pope, ANL
10:00-10:45 Python on HPC - Taylor Childers, ANL
10:45-11:00 Break
11:00-12:00 Deep Learning with Keras, Tensorflow, PyTorch, and Horovod on ThetaHuihuo Zheng, ANL
12:00-12:30 Overview of ADSP and Science Highlight: Deep Learning and Multimessenger AstrophysicsElise Jennings, ANL
12:30-1:15 ALCF Allocation Programs (working lunch) - Katherine Riley, ANL
1:15-2:00 Parallel I/O and Storage - Kevin Harms, ANL
2:00-2:45 Workflow Management with BalsamMisha Salim, ANL
2:45-3:30 DeepHyper - Prasanna Balaprakash and Romit Maulik, ANL
3:30-3:40 Break
3:40-5:30 Hands-on Session/Breakout Session on DeepHyper
5:30 Shuttle Departs from TCS to Guest House

Day 2 (Wednesday, October 2)

8:30 Shuttle Departs from Guest House to TCS, Bldg. 240
9:00-9:30 Best Practices for Deep Learning for Science, Bethany Lusch, ANL
9:30-10:15 

Success Stories of Science Applications Using Simulations, Machine Learning Tools and Data Science:

Ganesh Sivaraman, Ming Du, Wushi Dong

10:15-11:45 TAU Performance Analysis - Nick Chaimov, Paratools
11:45-12:00 Group Photo
12:00-1:00 Presentations from Attendees (working lunch)
1:00-1:45 Debugging and Profiling with DDT and MapRyan Hulguin, Arm 
1:45-2:30 Profiling Your Application with Intel VTune and AdvisorPaulius Velesko, Intel
2:30-3:15 Using OpenMP Effectively on Theta - Ron Green, Intel
3:15-3:30 Break
3:30-5:30 Hands-on Session/Breakout Session Tuning Performance with OpenMP
5:30 Shuttle Departs from TCS to Guest House

Day 3 (Thursday, October 3)

8:30 Shuttle Departs from Guest House to TCS, Bldg. 240
9:00-9:45 Using Containers on Theta - Murat Keceli, ANL
9:45-10:15 Aurora Exascale System Overview, Thomas Applencourt, ANL
10:15-10:30 Break
10:30-12:00 Hands-On/Breakout Session Profiling Deep Learning Frameworks
12:00-1:00 Presentations from Attendees (working lunch)
1:00-1:45 Visualizing Your Data - Joseph Insley and Silvio Rizzi, ANL
1:45-2:15 Using MPI Effectively on Theta - Sudheer Chunduri, ANL
2:15-2:30 Break
2:30-3:15 Getting Started with Spark on Theta - Xiao-Yong Jin, ANL
3:15-3:30 Director's Discretionary (DD) Program - Yasaman Ghadar, ANL
3:30-5:00 Hands-on Session
 

NOTE: No shuttle to Guest House. For assistance, visit registration desk.

LODGING

A block of rooms has been reserved at the Argonne Guest House at a rate of $91.50 + 6% tax.

Reservations may be made online or by phone (630) 739-6000.

When making a reservation, please specify the Argonne Contact: Monica White, mowhite@anl.gov, ALCF Division

Room block group name: SDL Workshop

Guest House website: https://www.anlgh.org

 

FOOD

http://tinyurl.com/eats-near-anl

 

Your Required Gate Pass/Badge

Argonne is a badged facility. A gate pass (issued as a clip-on ID badge) is required to enter Argonne the lab. Once you are approved to attend this event, we will arrange for a gate pass to be ready for you to pick up at the Argonne Information Center, building 224, located outside the main Argonne gate ("Northgate") on Northgate Road off Cass Avenue. The Argonne Information Center is open from 6:30 a.m. to 4:55 p.m. central. If you arrive after hours, your gate pass will be available from the guards at the guardhouse. If you have not arranged in advance for your driver to receive a gate pass, he/she will not be allowed to drive you into the laboratory. See the Getting to Argonne tab for instructions.

Visitors who have been issued a gate pass may enter the site for Laboratory-approved activities subject to the time and date restrictions on their gate pass. All visitors are required to present photo identification, such as a driver’s license or passport, to receive a gate pass. Non-U.S. citizens must also provide INS documents (visa, passports, I-20, I-797A and/or DS2019). In addition, all those operating a motor vehicle on site must have a valid driver’s license. Passes must be worn in plain view, on the upper part of the body, at all times while on site. Gate passes may be used only to gain site access for the person to whom the pass was issued. Visitors, except Argonne Guest House or Lodging residents, may not authorize site access or issuance of a gate pass for any other person.

Getting to Argonne

If you are not traveling to Argonne via private or rental car, please note that limousine and taxi drivers who are not US citizens may be unable to enter the lab. We strongly recommend you contact United Limousine for your ground transportation needs. As a contracted provider, United Limousine is familiar with Argonne security processes and will assure your driver meets the requirements to enter the laboratory.

  • United Limousine Reservations: 1-630-969-3865 (IMPORTANT: Be sure to notify United Limousine that you are traveling to Argonne National Laboratory.)
  • Airport Arrivals: 1-800-331-9037 or 1-630-969-3865

Argonne Leadership Computing Facility (ALCF)
Building 240
9700 S. Cass Avenue
Argonne, IL  60439
North Entrance to Argonne in Google Maps

To Argonne coming from O'Hare International Airport: Take I-294 south to I-55. Exit west on I-55 (toward St. Louis) and continue for about four miles to Cass Avenue. Exit south on Cass and turn right at the Argonne sign on Northgate Road, immediately south of I-55. Follow Northgate Road to the Argonne Information Center.

To Argonne coming from Midway Airport: Take Cicero Avenue north to I-55. Enter I-55 south and continue for about 14 miles to Cass Avenue. Exit south on Cass and turn right at the Argonne sign on Northgate Road, immediately south of I-55. Follow Northgate Road to the Argonne Information Center.

Speakers

Speaker/Staff  Expertise
Alexeev, Yuri: CPS, Principal Project Specialist, Quantum Computing, GAMESS
Chunduri, Sudheer: ALCF, Assistant Computer Scientist, MPI, Networking 
Ghadar, Yasaman: ALCF, Assistant Computational Scientist, LAMMPS, Training and Workshops
Harms, Kevin: ALCF, Lead, I/O Libraries & Benchmarks 
Hulguin, Ryan: ARM, Allinea DDT debugger and MAP performance profiler
Insley, Joseph: ALCF, Lead, Visualization & Data Analytics
Jennings, Elise: ALCF, Computer Scientist, Statistics, Machine Learning 
Jin, Xiao-Yong: CPS, Assistant Computer Scientist, Lattice QCD, Spark
Keceli, Murat: CPS, Assistant Computational Scientist, Valance Band Theory, Machine Learning
Knight, Christopher: CPS, Computational Scientist/Team Lead, DFT, Molecular Simulations, LAMMPS, Qbox, WEST
Kwack, JaeHyuk: ALCF, Assistant Computer Scientist, Performance Engineering
Li, Ying: CPS, Argonne Scholar, Molecular Mechanics 
Loy, Ray: ALCF, Lead for Training, Debuggers, and Math Libraries
Osborn, James: CPS, Computational Scientist/Team Lead, Floating-point Intrinsics, Low-level Communications, Nuclear & High Energy Physics, Sparse Linear Algebra
Papka, Michael: Deputy Associate Lab Director (Computing, Environment, and Life Sciences), Division Director (ALCF)
Ringenburg, Michael: Cray, Machine Learning including Spark, Alchemist
Rizzi, Silvio: ALCF, Assistant Computational Scientist, Visualization
Romero, Nichols: CPS, Principal Project Specialist, Computational Science, Electronic Structure of Condensed Matter, Programming Models
Salim, Misha: ALCF, Workflows, Balsam
Sivaraman, Ganesh: ALCF, Data Science, Machine Learning
Uram, Thomas: ALCF, Data Science, Workflows, Balsam
Vishwanath, Venkat: ALCF, Lead, Data Science, Machine Learning
White, Monica Orozco: ANL, Adminstrative Assistant
Luo, Ye: CPS, Assistant Scientific Applications Engineer, QMC

Job Submission

qsub SDL_Workshop q training

Workshop Machine Reservations 

theta.alcf.anl.gov (11.69 PF KNL XC-40)  
•9AM-6PM  1000 nodes   Queue: “training”
•6PM-12AM 2000 nodes  Queue: “training-pm”
cooley.alcf.anl.gov (x86+GPU viz/analysis cluster)
•9AM-6PM  50 nodes  Queue: “training”
 

Slides will be linked from agenda soon

—Recent uploads at: tinyurl.com/sdl-oct2019-slides
 

Online documentation