The Big Data Summit 2018, hosted by NERSC, will feature results from advanced Data Analytics and Management projects on the NERSC Cori system. We will feature ongoing projects being led by NERSC, Intel, Cray, and five Intel® Parallel Computing Centers (Intel® PCC)—represented by Oxford University, University of Liverpool, University of California – Berkeley, University of California – Davis, and New York University. The event will be comprised of plenary speakers, technical and invited talks, and a poster session. To register, please complete the form below.
This event is free and open to the public, but advance registration is required (registration deadline July 11, 2018). Attendees will cover their own meals (cafeteria available on site). Beverages will be provided throughout the day by NERSC.
Conference Date: July 18, 2018
Live Stream: A live stream of the event will be available at:
https://zoom.us/j/5104867752Agenda:
8:30-9:00 a.m.
Registration and Refreshments
9:00-9:45 a.m.
Welcome and Agenda Review; NERSC Update – Analytics and Management (Prabhat, NERSC)
9:45-10:15 a.m.
HPC and ML – The New Frontier of Compute (Victor Lee, Intel)
10:15-10:45 a.m.
Data Science and Machine Learning at Supercomputer Scale (Mike Ringenburg, Cray)
10:45-11:00 a.m.
Break
11:00-11:30 a.m.
Distributed Training of Generative Adversarial Networks for Fast Detector Simulation on Intel® Xeon® HPC Cluster (Vikram Saletore and Hans Pabst, Intel; Damian Podareanu and Valeriu Codreanu, SURFsara B.V.; Sofia Vallecorsa and Federico Carminati, CERN)
11:30-12:00 p.m.
Deep Probabilistic Models for Astronomy (Jeffrey Regier, University of California – Berkeley)
12:00 - 12:30 p.m.
Project DisCo: Physics-based Discovery of Coherent Structures in Spatiotemporal Systems (Adam Rupe, James P. Crutchfield, and Ryan G. James, University of California – Davis)
12:30-1:30 p.m.
Lunch
1:30 - 2:00 p.m.
Working Towards Distributed Inference Compilation at Scale (Frank Wood, University of British Columbia (formerly Oxford University))
2:00-2:30 p.m.
Neutrino Detection with Graph Neural Networks (Nicholas Choma and Joan Bruna, New York University)
2:30-3:00 p.m.
Machine Learning and Topological Data Analysis: Applications to Pattern Classification in Fluid and Climate Simulations (Vitaliy Kurlin and Grzegorz Muszynski, University of Liverpool)
3:00-3:30 p.m.
Deep Dive into TensorFlow: Single and Multi-Node Optimizations on Intel Xeon (Ashraf Bhuiyan, Intel; Steven Farrell, NERSC)
3:30-3:45 p.m.
Break
3:45 - 4:15 p.m.
High Productivity Languages – Python (Heidi Pan, Intel; Rollin Thomas, NERSC)
4:15 - 4:45 p.m.
CosmoFlow: Using Deep Learning to Learn the Universe at Scale (Debbie Bard, NERSC)
4:45 - 5:15 p.m.
Data Management and I/O Deep Dive (Quincey Koziol, NERSC)
5:15-5:30 p.m.
Closing Remarks (Prabhat and Quincey Koziol, NERSC)
5:30-7:00 p.m.
Poster Session and Refreshments
Venue: Lawrence Berkeley National Laboratory, 1 Cyclotron Rd, Berkeley, CA 94720 (Wang Hall, Building 59, Room 3101) Tel: 510.486.7612
Visitor Information:
https://docs.google.com/document/d/1ooS6B6T6zlrtVnfmxYXa66extFL2VhVSEXtdbREkPIk/edit?usp=sharingVisa Requirements and Questions: Contact IPCC.Program.Office@Intel.com attn: Shari Lawrence, Program Manager.
