Brookhaven Lab Staff to Participate at SC17
Staff will showcase their latest advances in high-performance computing, networking, storage, and analysis to the international supercomputing community
November 13, 2017
Staff from the Computational Science Initiative (CSI) and the Nuclear and Particle Physics Directorate at the U.S. Department of Energy’s Brookhaven National Laboratory will again be sharing their expertise with the worldwide supercomputing community at the annual SuperComputing (SC) conference, to be held November 12 through 17 at the Colorado Convention Center in Denver, Colorado. They will participate in several presentations, tutorials, workshops, demonstrations, and informal discussion groups, as outlined below.
Saturday, November 11 (Preconference)
Intel HPC Developer Conference Panels
“Industry Panel: New Paradigm in Computing: What Should Developers Know?” speakers include Barbara Chapman, CSI Computer Science and Mathematics Department; 12:15–12:50 pm, Grand Ballroom
Sunday, November 12
Tutorials
“OpenMP Common Core: A “Hands-On” Exploration,” presenters include Barbara Chapman, CSI Computer Science and Mathematics Department; 8:30 am–5:00 pm, Room 303
Paper Presentations
“Toward Common Components for Open Workflow Systems (Workshop on Open Source Supercomputing),” authors include Shantenu Jha, CSI Center for Data-Driven Discovery; 11:30 am–12:00 pm, Room 506
“Why Wait? Let’s Start Computing While Data is Still on the Wire (Innovating the Network for Data-Intensive Science Workshop),” Shilpi Bhattacharyya, Dimitrios Katramatos, and Shinjae Yoo, CSI Computer Science and Mathematics Department, 3:40–4:00 pm, Room 607
Workshops
“Use Cases of Computational Reproducibility for Scientific Workflows at Exascale (Computational Reproducibility at Exascale 2017, CRE2017),” presenter is Line Pouchard, CSI Center for Data-Driven Discovery; 4:20 pm, Room 702
“Designing Workflow Management Systems Using Building Blocks (Workflows in Support of Large-Scale Science, WORKS 2017 Workshop),” authors include Shantenu Jha, CSI Center for Data-Driven Discovery; 4:30–4:50 pm, Room 501
Monday, November 13
Workshops
“Energy Efficient Supercomputing (E2SC),” organizers include Adolfy Hoisie, CSI Computing for National Security Department; 9:00 am–5:30 pm, Room 506
Presentations
“Evaluating On-Node GPU Interconnects for Deep Learning Workloads (The 8th International Workshop on Performance Modeling, Benchmarking, and Simulation of High-Performance Computer Systems [PMBS17]),” authors include Adolfy Hoisie, CSI Computing for National Security Department; 9:01–9:30 am, Rooms 704–706
“Benchmarking and Evaluating Unified Memory for OpenMP GPU Offloading (Fourth Workshop on the LLVM Compiler Infrastructure in HPC),” authors/presenters include Barbara Chapman, Martin Kong, and Lingda Li, CSI Computer Science and Mathematics Department; 11:00–11:30 am, Room 710
“Developing an OpenMP Runtime for UVM-Capable GPUs,” authors include Lingda Li, CSI Computer Science and Mathematics Department; 5:07–5:14 pm, Room 710
Tuesday, November 14
Featured Talks in the DOE Booth
“BigPanDA Project,” presenter is Alexei Klimentov, Brookhaven Lab Physics Department; 4:45 pm, Exhibit Floor, Booth 613
Demos in the DOE Booth
“BigPanDA @Titan,” presenter is Sergey Panitkin, Brookhaven Lab Physics Department; 12:00 pm, Exhibit Floor, Booth 613
Roundtable in the DOE Booth
“BigPanDA (DOE ASCR funded project),” convener is Alexei Klimentov, Brookhaven Lab Physics Department; 5:30pm, Exhibit Floor, Booth 613
Poster Presentations
“A19: Performance Analysis of a Parallelized Restricted Boltzmann Machine Artificial Neural Network Using OpenACC Framework and TAU Profiling System,” author is Abhishek Kumar, CSI Computer Science and Mathematics Department; 5:15–7:00 pm, Four Seasons Ballroom
“P34: GPU Acceleration for the Impurity Solver in GW+DMFT Packages,” authors are Nicholas D’Imperio, Patrick Semon, Kwangmin Yu, CSI Computational Science Laboratory; 5:15–7:00 pm, Four Seasons Ballroom
Wednesday, November 15
Talk
“ECP SOLLVE: Taking OpenMP to Exascale,” speaker is Martin Kong, CSI Computer Science and Mathematics Department; 11:15 am, Exhibit Floor, OpenMP Booth 1246
Demos in the DOE Booth
“ColorMap-ND: Mapping Multivariate Data to Color,” presenter is Klaus Mueller, CSI Center for Data-Driven Discovery; 1:00 pm, Exhibit Floor, Booth 613, Demo Station 1
“Visual Analytics for Heterogeneous Scientific Data Exploration,” presenter is Wei Xu, CSI Center for Data-Driven Discovery; 3:00 pm, Exhibit Floor, Both 613, Demo Station 2
“OpenMP 5.0 Memory Management and Unified Memory Space,” presenter is Lingda Li, CSI Computer Science and Mathematics Department; 4:00 pm, Exhibit Floor, Booth 613, Demo Station 1
Paper Presentations
“Representative Paths Analysis,” authors include Adolfy Hoisie, CSI Computing for National Security Department; 2:00–2:30 pm, Rooms 402—404
Featured Talks in the DOE Booth
“OpenMP for the Long Haul,” presenter is Barbara Chapman, CSI Computer Science and Mathematics Department; 2:30 pm, Exhibit Floor, Booth 613
Birds-of-a-Feather Sessions
“OpenSHMEM in the Era of Exascale,” additional session leaders include Barbara Chapman, CSI Computer Science and Mathematics Department; 5:15–7:00 pm, Rooms 501–502
“OpenMP,” panelist is Barbara Chapman, CSI Computer Science and Mathematics Department, 6:15 pm, Rooms 205–207
Thursday, November 16
Demos in the DOE Booth
“BigPanDA @Titan,” presenter is Sergey Panitkin, Brookhaven Lab Physics Department; 12:00 pm, Exhibit Floor, Booth 613
“ColorMap-ND: Mapping Multivariate Data to Color,” presenter is Klaus Mueller, CSI Center for Data-Driven Discovery; 1:00 pm, Exhibit Floor 613, Demo Station 2
Roundtable at the DOE Booth
“BigPanDA (DOE ASCR funded project),” convener is Sergey Panitkin, Brookhaven Lab Physics Department; 10:00 am, Exhibit Floor, Booth 613
Friday, November 17
Workshop
“High-Throughput Binding Affinity Calculations at Extreme Scales (Computational Approaches for Cancer Workshop),” authors include Shantenu Jha, CSI Center for Data-Driven Discovery; 11:00–11:30 am, Room 404
CSI integrates computer science, applied mathematics, computational science, and domain science expertise and investments across Brookhaven Lab—including its flagship facilities that attract thousands of scientific users each year—to tackle the big data challenges at the frontiers of scientific discovery.
Brookhaven National Laboratory is supported by the Office of Science of the U.S. Department of Energy. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States, and is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.
2017-12620 | INT/EXT | Newsroom
