IEEE Cluster 2019 Program


Overview | By Date | By Event Type | By Room | Author Index



Keynote
Tuesday 9:00am-10:00am
Ambassador/Registry
Keynote
Keynote 1
Chair: Martin Schulz (Technical University of Munich)
RAPIDS: Open Source Python Data Science with GPU Acceleration and Dask (Joe Eaton)
Joe Eaton (Nvidia)
Biography
Abstract
Wednesday 9:00am-10:00am
Ambassador/Registry
Keynote
Keynote 2
Chair: Patrick McCormick (Los Alamos National Laboratory)
AI@Edge (Pete Beckman)
Pete Beckman (Argonne National Laboratory)
Biography
Abstract
Thursday 9:00am-10:00am
Ambassador/Registry
Keynote
Keynote 3
Chair: Patrick G. Bridges (University of New Mexico)
Big Data Spatiotemporal Analytics - Trends, Characteristics and Applications (Sangmi Lee Pallickara)
Sangmi Pallickara (Colorado State University)
Biography
Abstract

Other
Monday 7:00am-9:00am
Foyer
Other
Registration and Badge Pickup
Tuesday 7:00am-8:45am
Foyer
Other
Registration and Badge Pickup
Tuesday 10:30am-11:00am
Wurlitzer
Other
Break
Tuesday 12:30pm-2:00pm
Other
Other
Lunch (on your own)
Tuesday 3:30pm-4:00pm
Wurlitzer
Other
Break
Wednesday 7:00am-8:45am
Foyer
Other
Registration and Badge Pickup
Wednesday 10:30am-11:00am
Wurlitzer
Other
Break
Wednesday 12:30pm-2:00pm
Other
Other
Lunch (on your own)
Wednesday 3:30pm-4:00pm
Wurlitzer
Other
Break
Wednesday 6:30pm-10:00pm
Roxy
Other
Conference Banquet
Conference Dinner
Abstract
Thursday 7:00am-8:45am
Foyer
Other
Registration and Badge Pickup
Thursday 10:30am-11:00am
Wurlitzer
Other
Break

Papers
Tuesday 10:00am-10:30am
Ambassador/Registry
Paper
Top 3 Papers of Cluster 2019 (1/3)
Chair: Martin Schulz (Technical University of Munich)
Evaluating Burst Buffer Placement in HPC Systems
Best Paper
Harsh Khetawat (NCSU), Christopher Zimmer (ORNL), Frank Mueller (NCSU), Scott Atchley and Sudharshan Vazhkudai (ORNL), and Misbah Mubarak (ANL)
Abstract
Tuesday 11:00am-12:30pm
Ambassador/Registry
Paper
Deep Learning 1
Chair: Judy Qiu (Indiana University)
Efficient User-Level Storage Disaggregation for Deep Learning
Yue Zhu, Weikuan Yu, and Bing Jiao (Florida State University); Kathryn Mohror and Adam Moody (Lawrence Livermore National Laboratory); and Fahim Chowdhury (Florida State University)
Abstract
FluentPS: A Parameter Server Design with Low-frequency Synchronization for Distributed Deep Learning
Xin Yao, Xueyu Wu, and Cho-Li Wang (HKU)
Abstract
Performance Characterization of DNN Training using TensorFlow and PyTorch on Modern Clusters
Arpan Jain, Ammar Ahmad Awan, Quentin Anthony, Hari Subramoni, and Dhabaleswar K. Panda (The Ohio State University)
Abstract
Tuesday 11:00am-12:30pm
Regal
Paper
Parallel Applications Using Alternate Models
Chair: David Boehme (Lawrence Livermore National Laboratory)
Leveraging Task-Based Polar Decomposition Using PARSEC on Massively Parallel Systems
Dalal Sukkari (KAUST), Mathieu Faverge (INRIA), and Hatem Ltaief and David Keyes (KAUST)
Abstract
Engineering a Distributed Histogram Sort
Roger Kowalewski, Pascal Jungblut, and Karl Fuerlinger (LMU Munich)
Abstract
Asynchronous Task-Based Execution of the Reverse Time Migration for the Oil and Gas Industry
Amani Alonazi and Hatem Ltaief (KAUST); Issam Said (NVIDIA); Samuel Thibault (University of Bordeaux, LaBRI – INRIA Bordeaux Sud-Ouest); and David Keyes (KAUST)
Abstract
Tuesday 2:00pm-3:30pm
Ambassador/Registry
Paper
Deep Learning 2
Chair: Hatem Ltaief (KAUST)
A Quantitative Study of Deep Learning Training on Heterogeneous Supercomputers
Jingoo Han (Virginia Tech), Luna Xu (IBM Research), M. Mustafa Rafique (Rochester Institute of Technology), Ali R. Butt (Virginia Tech), and Seung-Hwan Lim (Oak Ridge National Laboratory)
Abstract
Parallelizing Training of Deep Generative Models on Massive Scientific Datasets
Sam Ade Jacobs, Brian Van Essen, Tim Moon, Jae Seung Yeom, David Hysom, Brian Spears, Rushil Anirudh, Jayaraman Thiagaranjan, Shusen Liu, Jim Gaffney, Peer-Timo Bremer, Tom Benson, Peter Robinson, and Luc Peterson (Lawrence Livermore National Lab)
Abstract
Quantifying the Impact of Memory Errors in Deep Learning
Zhao Zhang, Lei Huang, Ruizhu Huang, and Weijia Xu (TACC) and Daniel S. Katz (University of Illinois)
Abstract
Tuesday 2:00pm-3:30pm
Regal
Paper
Workflows
Chair: Sameer Shende (University of Oregon; ParaTools, Inc.)
NORNS: Extending Slurm to Support Data-Driven Workflows through Asynchronous Data Staging
Alberto Miranda (Barcelona Supercomputing Center); Adrian Jackson (EPCC, The University of Edinburgh); Tommaso Tocci (Barcelona Supercomputing Center); Iakovos Panourgias (EPCC, The University of Edinburgh); and Ramon Nou (Barcelona Supercomputing Center)
Abstract
Leveraging Machine Learning for Anticipatory Data Delivery in Extreme Scale In-situ Workflows
Pradeep Subedi, Philip E. Davis, and Manish Parashar (Rutgers University)
Abstract
Harmony: An Approach for Geo-distributed Processing of Big-Data Applications
Han Zhang (National University of Singapore), Lavanya Ramapantulu (International Institute of Information Technology), and Yong Meng Teo (National University of Singapore)
Abstract
Tuesday 4:00pm-5:00pm
Ambassador/Registry
Paper
Machine Learning
Chair: Ann Gentile (Sandia National Laboratories)
HarpGBDT: Optimizing Gradient Boosting Decision Tree for Parallel Efficiency
Bo Peng and Judy Qiu (Indiana University)
Abstract
Training Google Neural Machine Translation on an Intel CPU Cluster
Dhiraj Kalamkar, Kunal Banerjee, Sudarshan Srinivasan, Srinivas Sridharan, Evangelos Georganas, Mikhail Smorkalov, Cong Xu, and Alexander Heinecke (Intel)
Abstract
Tuesday 4:00pm-5:00pm
Regal
Paper
Clustering
Chair: Ron Brightwell (Sandia National Laboratories)
MuDBSCAN: An Exact Scalable DBSCAN Algorithm for Big Data Exploiting Spatial Locality
Aditya Sarma, Poonam Goyal, Sonal Kumari, Anand Wani, Jagat Sesh Challa, Saiyedul Islam, and Navneet Goyal (Birla Institute of Technology & Science, Pilani)
Abstract
Wednesday 10:00am-10:30am
Ambassador/Registry
Paper
Top 3 Papers of Cluster 2019 (2/3)
Chair: Patrick McCormick (Los Alamos National Laboratory)
Algorithm-Based Fault Tolerance for Parallel Stencil Computations
Best Paper
Aurélien Cavelan and Florina M. Ciorba (University of Basel, Swizterland)
Abstract
Wednesday 11:00am-12:30pm
Ambassador/Registry
Paper
Message Passing
Chair: Felix Wolf (TU Darmstadt)
MPI Sessions: Evaluation of an Implementation in Open MPI
Nathan Hjelm (Google Inc.); Howard Pritchard and Samuel K. Gutiérrez (Los Alamos National Laboratory); Daniel J. Holmes (EPCC, The University of Edinburgh); Ralph Castain (Intel); and Anthony Skjellum (University of Tennessee at Chattanooga)
Abstract
Give MPI Threading a Fair Chance: A Study of Multithreaded MPI Designs
Thananon Patinyasakdikul, David Eberius, and George Bosilca (University of Tennessee) and Nathan Hjelm (University of New Mexico)
Abstract
Fast and Faithful Performance Prediction of MPI Applications: the HPL Case Study
Tom Cornebize (Université Grenoble Alpes, French Institute for Research in Computer Science and Automation (INRIA)); Arnaud Legrand (National Center for Scientific Research (CNRS), French Institute for Research in Computer Science and Automation (INRIA)); and Franz Christian Heinrich (French Institute for Research in Computer Science and Automation (INRIA))
Abstract
Wednesday 11:00am-12:30pm
Regal
Paper
Data Centers and Clouds
Chair: Kevin Pedretti (Sandia National Laboratories)
MBECN: Enabling ECN with Micro-burst Traffic in Multi-queue Data Center
Kexi Kang, Jinghui Zhang, Jiahui Jin, Dian Shen, and Junzhou Luo (Southeast University); Wenxin Li (Hong Kong University of Science and Technology); and Zhiang Wu (Nanjing University of Finance and Economics)
Abstract
Large-Scale Analysis of the Docker Hub Dataset
Nannan Zhao (Virginia Tech); Vasily Tarasov (IBM Research—Almaden); Hadeel Albahar (Virginia Tech); Ali Anwar, Lukas Rupprecht, Dimitrios Skourtis, and Amit S. Warke (IBM Research—Almaden); Mohamed Mohamed (Apple); and Ali R. Butt (Virginia Tech)
Abstract
DP_Greedy: A Two-Phase Caching Algorithm for Mobile Cloud Services
Dong Huang, Xiaopeng Fan, and Yang Wang (Shenzhen Institutes of Advanced Technology); Shuibing He (Zhejiang University); and Chengzhong Xu (University of Macau)
Abstract
Wednesday 2:00pm-3:30pm
Ambassador/Registry
Paper
Cluster Communication
Chair: Scott Levy (Sandia National Laboratories)
X-RDMA: Effective RDMA Middleware in Large-scale Production Environments
Teng Ma (Tsinghua University, Alibaba); Tao Ma, Zhuo Song, Jingxuan Li, and Huaixin Chang (Alibaba); Kang Chen (Tsinghua University); Hai Jiang (Arkansas State University); and Yongwei Wu (Tsinghua University)
Abstract
Propagation and Decay of Injected One-Off Delays on Clusters: A Case Study
Ayesha Afzal, Georg Hager, and Gerhard Wellein (Friedrich-Alexander University Erlangen-Nürnberg)
Abstract
An Empirical Study of Cryptographic Libraries for MPI Communications
Abu Naser, Mohsen Gavahi, Cong Wu, Viet Tung Hoang, Zhi Wang, and Xin Yuan (Florida State University)
Abstract
Wednesday 2:00pm-3:30pm
Regal
Paper
Efficient Storage
Chair: Kathryn Mohror (LLNL)
RE-Store: Reliable and Efficient KV-Store with Erasure Coding and Replication
Yuzhe Li, Jiang Zhou, and Weiping Wang (Institute of Information Engineering, Chinese Academy of Sciences) and Yong Chen (Texas Tech University)
Abstract
Compact Filter Structures for Fast Data Partitioning
Qing Zheng, Charles Cranor, Ankush Jain, Gregory Ganger, Garth Gibson, and George Amvrosiadis (Carnegie Mellon University) and Bradley Settlemyer and Gary Grider (Los Alamos National Lab)
Abstract
Building Reliable High-Performance Storage Systems: An Empirical and Analytical Study
Zhi Qiao (University of North Texas; USRC, LANL); Song Fu (University of North Texas); and Hsing-Bung Chen and Bradley Settlemyer (Los Alamos National Laboratory)
Abstract
Wednesday 4:00pm-5:15pm
Regal
Paper
Compression
Chair: Christian Engelmann (Oak Ridge National Laboratory)
Analyzing the Impact of Lossy Compressor Variability on Checkpointing Scientific Simulations (short paper)
Pavlo D. Triantafyllides, Tasmia Reza, and Jon C. Calhoun (Clemson University)
Abstract
Improving Performance of Data Dumping with Lossy Compression for Scientific Simulation
Xin Liang (UC, Riverside); Sheng Di (Argonne National Laboratory); Dingwen Tao (the University of Alabama); Sihuan Li (UC, Riverside); Bogdan Nicolae (Argonne National Laboratory); Zizhong Chen (UC, Riverside); and Franck Cappello (Argonne National Laboratory)
Abstract
Efficient Distributed Graph Analytics using Triply Compressed Sparse Format
Mohammad Hasanzadeh Mofrad and Rami Melhem (University of Pittsburgh) and Yousuf Ahmad and Mohammad Hammoud (Carnegie Mellon University in Qatar)
Abstract
Wednesday 4:00pm-5:30pm
Ambassador/Registry
Paper
Resource Allocation
Chair: Eishi Arima (The University of Tokyo)
SMQoS: Improving Utilization and Power Efficiency with QoS Awareness on GPUs (short paper)
Qingxiao Sun, Yi Liu, Hailong Yang, Zhongzhi Luan, and Depei Qian (Beihang University)
Abstract
Mitigating Inter-Job Interference via Process-Level Quality-of-Service (short paper)
Lee Savoie and David Lowenthal (University of Arizona), Bronis de Supinski and Kathryn Mohror (Lawrence Livermore National Laboratory), and Nikhil Jain (Nvidia)
Abstract
Kube-Knots: Resource Harvesting through Dynamic Container Orchestration in GPU-based Datacenters
Prashanth Thinakaran and Jashwant Raj Gunasekaran (Penn State), Bikash Sharma (Facebook), and Mahmut Kandemir and Chita Das (Penn State)
Abstract
Scheduling Independent Stochastic Tasks on Heterogeneous Cloud Platforms
Yiqin Gao (ENS Lyon); Louis-Claude Canon (Univ. Franche Comté); Yves Robert (ENS Lyon, Univ. Tenn. Knoxville); and Frédéric Vivien (Inria)
Abstract
Thursday 10:00am-10:30am
Ambassador/Registry
Paper
Top 3 Papers of Cluster 2019 (3/3)
Chair: Patrick G. Bridges (University of New Mexico)
STASH : Fast Hierarchical Aggregation Queries for Effective Visual Spatiotemporal Explorations
Best Paper
Saptashwa Mitra, Paahuni Khandelwal, Shrideep Pallickara, and Sangmi Lee Pallickara (Colorado State University)
Abstract
Thursday 11:00am-12:30pm
Ambassador/Registry
Paper
Tools and Optimization
Chair: Bronis R. de Supinski (Lawrence Livermore National Laboratory)
DiffTrace: Efficient Whole-Program Trace Analysis and Diffing for Debugging
Saeed Taheri and Ian Briggs (University of Utah), Martin Burtscher (Texas State University), and Ganesh Gopalakrishnan (University of Utah)
Abstract
FSMonitor: Scalable File System Monitoring for Arbitrary Storage Systems
Arnab K. Paul (Virginia Tech); Ryan Chard (Argonne National Laboratory); Kyle Chard and Steven Tuecke (University of Chicago); Ali R. Butt (Virginia Tech); and Ian Foster (Argonne National Laboratory, University of Chicago)
Abstract
On the Benefits of Anticipating Load Imbalance for Performance Optimization of Parallel Applications
Anthony Boulmier (University of Geneva), Franck Raynaud (Unversity of Geneva), Nabil Abdennadher (HES-SO), and Bastien Chopard (Unversity of Geneva)
Abstract
Thursday 11:00am-12:30pm
Regal
Paper
Applications
Chair: Ryan Grant (Sandia National Labs, University of New Mexico)
Multi-physics simulations of particle tracking in arterial geometries with a scalable moving window algorithm
Gregory J. Herschlag (Duke University), John Gounley (Oak Ridge National Laboratory), Sayan Roychowdhury (Duke University), Erik W. Draeger (Lawrence Livermore National Laboratory), and Amanda Randles (Duke University)
Abstract
Fast and Scalable Implementations of Influence Maximization Algorithms
Marco Minutoli (Pacific Northwest National Laboratory, Washington State University); Mahantesh Halappanavar (Pacific Northwest National Laboratory); Ananth Kalyanaraman (Washington State University); and Arun Sathanur, Ryan Mcclure, and Jason McDermott (Pacific Northwest National Laboratory)
Abstract
Scalable, High-Order Continuity Across Block Boundaries of Functional Approximations Computed in Parallel
Iulian Grindeanu, Tom Peterka, Vijay Mahadevan, and Youssef S. Nashed (Argonne National Laboratory)
Abstract

Plenary
Tuesday 8:45am-9:00am
Ambassador/Registry
Plenary
Cluster 2019 Opening
Chair: Ron Brightwell (Sandia National Laboratories); Patrick G. Bridges (University of New Mexico); Martin Schulz (Technical University of Munich); Patrick McCormick (Los Alamos National Laboratory)
Wednesday 8:45am-9:00am
Ambassador/Registry
Plenary
Announcements
Chair: Ron Brightwell (Sandia National Laboratories); Patrick G. Bridges (University of New Mexico)
Thursday 8:45am-9:00am
Ambassador/Registry
Plenary
Cluster 2020 Presentation
Thursday 12:30pm-12:45pm
Ambassador/Registry
Plenary
Closing Remarks
Chair: Ron Brightwell (Sandia National Laboratories); Patrick G. Bridges (University of New Mexico)

Posters
Tuesday 5:00pm-5:30pm
Ambassador/Registry
Poster
Poster Blitz
Tuesday 5:30pm-8:00pm
Wurlitzer
Poster
Poster Session and Reception

Workshops
Monday 9:00am-3:30pm
Regal
Workshop
Workshop on Monitoring and Analysis for High Performance Computing Systems Plus Applications
Monday 9:00am-6:00pm
Ambassador/Registry
Workshop
First Extreme-scale Scientific Software Stack Forum (E4S Forum)

Created 2019-9-26 12:7