conference-talk

Introduction to Deep Learning for science (single-node and multi-node training)

Date and Time: 
Thursday, April 11th 2019
Location: 
CG North Auditorium
Speaker: 
Steve Farrell

Deep learning is rapidly and fundamentally transforming the way science and industry use data to solve problems. Deep neural network models have been shown to be powerful tools for extracting insights from data across a large number of domains. As these models grow in complexity to solve increasingly challenging problems with larger and larger datasets, the need for scalable methods and software to train them grows accordingly.

Speaker Description: 

Steve Farrell is a Machine Learning Engineer at the NERSC supercomputing center at Lawrence Berkeley National Laboratory. In this role he supports the ML needs of 7000 users across a wide range of scientific domains, and the ML software stack. He collaborates with science teams to perform applied ML research.

Event Category:

Deep Learning for Science: Capabilities and Challenges for Transforming Scientific Workflows

Date and Time: 
KEYNOTE - Tuesday, April 9th 2019
Location: 
CG auditorium
Speaker: 
Steve Farrell

Modern science is getting bigger in every way: the scope of the problems, the size of the datasets, and the required computing power. Massive undertakings such as the Large Hadron Collider and the Large Synoptic Survey Telescope as well as major looming challenges such as global climate change are pushing science past its limits. Deep Learning, a methodology that has seen tremendous success in industry applications, is poised to transform scientific workflows.

Speaker Description: 

Steve Farrell is a Machine Learning Engineer at the NERSC supercomputing center at Lawrence Berkeley National Laboratory. In this role he supports the ML needs of 7000 users across a wide range of scientific domains, and the ML software stack. He collaborates with science teams to perform applied ML research.

Event Category:

Video recorded: 

The slides are avaialble here

Uncertainty Quantification with Analog Ensemble at Scale

Date and Time: 
Tuesday April 9th 2019
Location: 
CG Auditorium
Speaker: 
Weiming Hu

Model uncertainty estimation using the ensemble approach poses a large computation requirement. This is because ensembles are usually generated form multi-model and multi-simulation with slightly perturbed initialization.

Speaker Description: 

Weiming Hu is a Ph.D. student of Prof. Guido Cervone at Penn State University in the Dept. of Geography focusing on computational algorithms, and numerical weather prediction; Guido Cervone is the Associate Professor at Dept. of Geography and the Associate Director of Institue for CyberScience at Penn State University.

He can be reached at weiming@psu.edu if you have any questions.

Event Category:

Building Scalable NVIDIA GPU-Based clusters for HPC and Deep Learning

Date and Time: 
Monday April 8th 2019
Location: 
CG Auditorium
Speaker: 
Craig Tierney

Deep Learning model complexity and training data volume continue to grow rapidly. Training with a single-GPU, or even a single-node of GPUs, is often too slow for the iterative nature of model development and optimization. In this talk, we will discuss several aspects of building scalable GPU-based clusters for improving training time including scaling training to multiple-nodes, optimizing inter-node collective operations, and optimizing the data-cache hierarchy.

Speaker Description: 

Craig Tierney is a Senior Solution Architect at NVIDIA supporting high performance computing (HPC) and deep learning (DL). His focus includes the architecture of GPU based systems to maximize HPC and DL performance and scalability. Prior to joining NVIDIA, Craig spent over 15 years providing high performance computing architecture and computational science support to NOAA and several other government and educational organizations including DOE, DOD, NASA and Stanford University. Craig holds a Ph.D. in Aerospace Engineering Sciences from the University of Colorado at Boulder.

Event Category:

Exploiting Machine Learning Hardware Features to do HPC Work

Date and Time: 
April 8th @ 1:00 pm
Speaker: 
Greg Henry

Many modern processors, including features we’ll describe with upcoming Intel processors, provide acceleration on machine learning applications. But how can some of these features be leveraged for modern HPC?

This talk will review theory, performance and future research for using instructions and techniques originally developed with only machine learning in mind to do HPC faster and yet still maintain accuracy and convergence properties.

Speaker Description: 

Bio: -

Event Category:

How to Boost the Performance of HPC/AI Applications Using MVAPICH2 Library?

Date and Time: 
April 12th 2019, 9:00 am - 12:00 pm
Location: 
CG Center Auditorium
Speaker: 
DK Panda, Hari Subramoni

The tutorial will start with an overview of the MVAPICH2 libraries and their features. Next, we will focus on installation guidelines, runtime optimizations and tuning flexibility in-depth. An overview of configuration and debugging support in MVAPICH2 libraries will be presented. High-performance support for GPU-enabled clusters in MVAPICH2-GDR and many-core systems in MVAPICH2-X will be presented. The impact on the performance of the various features and optimization techniques will be discussed in an integrated fashion.

Speaker Description: 

Dr. Dhabaleswar K. (DK) Panda is a Professor and University Distinguished Scholar of Computer Science at the Ohio State University. He obtained his Ph.D. in computer engineering from the University of Southern California. His research interests include parallel computer architecture, high performance computing, communication protocols, files systems, network-based computing, and Quality of Service. He has published over 400 papers in major journals and international conferences related to these research areas. Dr. Panda and his research group members have been doing extensive research on modern networking technologies including InfiniBand, HSE and RDMA over Converged Enhanced Ethernet (RoCE). His research group is currently collaborating with National Laboratories and leading InfiniBand and 10GigE/iWARP companies on designing various subsystems of next generation high-end systems. The MVAPICH2 (High Performance MPI over InfiniBand, iWARP and RoCE) open-source software package, developed by his research group, are currently being used by more than 2,950 organizations worldwide (in 86 countries). This software has enabled several InfiniBand clusters (including the 1st one) to get into the latest TOP500 ranking. More than 507,000 downloads of these libraries have taken place from the project’s site. These software packages are also available with the stacks for network vendors (InfiniBand and iWARP), server vendors and Linux distributors. The RDMA-enabled Apache Hadoop, Spark and Memcached packages, consisting of acceleration for HDFS, MapReduce, RPC, Spark and Memcached, are publicly available from High-Performance Big Data (HiBD) project site: http://hibd.cse.ohio-state.edu. These packages are currently being used by more than 295 organizations in 35 countries. More than 28,450 downloads have taken place from the project’s site. The group has also been focusing on co-designing Deep Learning Frameworks and MPI Libraries. High-performance and scalable versions of the Caffe and TensorFlow frameworks are available from High-Performance Deep Learning (HiDL) Project site: site: http://hidl.cse.ohio-state.edu. Dr. Panda’s research is supported by funding from US National Science Foundation, US Department of Energy, and several industries including Intel, Cisco, SUN, Mellanox, Microsoft, QLogic, NVIDIA and NetApp. He is an IEEE Fellow and a member of ACM. More details about Dr. Panda, including a comprehensive CV and publications are available at http://web.cse.ohio-state.edu/~panda.2/.

 

Dr. Hari Subramoni is a research scientist in the Department of Computer Science and Engineering at the Ohio State University, USA, since September 2015. His current research interests include high performance interconnects and protocols, parallel computer architecture, network-based computing, exascale computing, network topology aware computing, QoS, power-aware LAN-WAN communication, fault tolerance, virtualization, big data and cloud computing. He has published over 70 papers in international journals and conferences related to these research areas. He has been actively involved in various professional activities in academic journals and conferences. Dr. Subramoni is doing research on the design and development of MVAPICH2 (High Performance MPI over InfiniBand, iWARP and RoCE) and MVAPICH2-X (Hybrid MPI and PGAS (OpenSHMEM, UPC and CAF)) software packages. He is a member of IEEE. More details about Dr. Subramoni are available at http://web.cse.ohio-state.edu/~subramoni.1/.

Event Category:

High Performance Distributed Deep Learning

Date and Time: 
April 12th 2019, 1:00 pm - 4:00 pm
Location: 
CG Center Auditorium
Speaker: 
DK Panda, Hari Subramoni, and Ammar Awan

The current wave of advances in Deep Learning (DL) has led to many exciting challenges and opportunities for Computer Science and Artificial Intelligence researchers alike. Modern DL frameworks like Caffe2, TensorFlow, Cognitive Toolkit (CNTK), PyTorch, and several others have emerged that offer ease of use and flexibility to describe, train, and deploy various types of Deep Neural Networks (DNNs). In this tutorial, we will provide an overview of interesting trends in DNN design and how cutting-edge hardware architectures are playing a key role in moving the field forward.

Speaker Description: 

Dr. Dhabaleswar K. (DK) Panda is a Professor and University Distinguished Scholar of Computer Science at the Ohio State University. He obtained his Ph.D. in computer engineering from the University of Southern California. His research interests include parallel computer architecture, high performance computing, communication protocols, files systems, network-based computing, and Quality of Service. He has published over 400 papers in major journals and international conferences related to these research areas. Dr. Panda and his research group members have been doing extensive research on modern networking technologies including InfiniBand, HSE and RDMA over Converged Enhanced Ethernet (RoCE). His research group is currently collaborating with National Laboratories and leading InfiniBand and 10GigE/iWARP companies on designing various subsystems of next generation high-end systems. The MVAPICH2 (High Performance MPI over InfiniBand, iWARP and RoCE) open-source software package, developed by his research group, are currently being used by more than 2,950 organizations worldwide (in 86 countries). This software has enabled several InfiniBand clusters (including the 1st one) to get into the latest TOP500 ranking. More than 507,000 downloads of these libraries have taken place from the project’s site. These software packages are also available with the stacks for network vendors (InfiniBand and iWARP), server vendors and Linux distributors. The RDMA-enabled Apache Hadoop, Spark and Memcached packages, consisting of acceleration for HDFS, MapReduce, RPC, Spark and Memcached, are publicly available from High-Performance Big Data (HiBD) project site: http://hibd.cse.ohio-state.edu. These packages are currently being used by more than 295 organizations in 35 countries. More than 28,450 downloads have taken place from the project’s site. The group has also been focusing on co-designing Deep Learning Frameworks and MPI Libraries. High-performance and scalable versions of the Caffe and TensorFlow frameworks are available from High-Performance Deep Learning (HiDL) Project site: site: http://hidl.cse.ohio-state.edu. Dr. Panda’s research is supported by funding from US National Science Foundation, US Department of Energy, and several industries including Intel, Cisco, SUN, Mellanox, Microsoft, QLogic, NVIDIA and NetApp. He is an IEEE Fellow and a member of ACM. More details about Dr. Panda, including a comprehensive CV and publications are available at http://web.cse.ohio-state.edu/~panda.2/.

 

Ammar Ahmad Awan received his B.S. and M.S. degrees in Computer Science and Engineering from National University of Science and Technology (NUST), Pakistan and Kyung Hee University (KHU), South Korea, respectively. Currently, Ammar is working towards his Ph.D. degree in Computer Science and Engineering at The Ohio State University. His current research focus lies at the intersection of High Performance Computing (HPC) libraries and Deep Learning (DL) frameworks. He previously worked on a Java-based Message Passing Interface (MPI) and nested parallelism with OpenMP and MPI for scientific applications. He has published 14 papers in conferences and journals related to these research areas. He actively contributes to various projects like MVAPICH2-GDR (High Performance MPI for GPU clusters, OMB (OSU Micro Benchmarks), and HiDL (High Performance Deep Learning). He is the lead author of the OSU-Caffe framework (part of HiDL project) that allows efficient distributed training of Deep Neural Networks.

 

Dr. Hari Subramoni is a research scientist in the Department of Computer Science and Engineering at the Ohio State University, USA, since September 2015. His current research interests include high performance interconnects and protocols, parallel computer architecture, network-based computing, exascale computing, network topology aware computing, QoS, power-aware LAN-WAN communication, fault tolerance, virtualization, big data and cloud computing. He has published over 70 papers in international journals and conferences related to these research areas. He has been actively involved in various professional activities in academic journals and conferences. Dr. Subramoni is doing research on the design and development of MVAPICH2 (High Performance MPI over InfiniBand, iWARP and RoCE) and MVAPICH2-X (Hybrid MPI and PGAS (OpenSHMEM, UPC and CAF)) software packages. He is a member of IEEE. More details about Dr. Subramoni are available at http://web.cse.ohio-state.edu/~subramoni.1/.

Event Category:

Bare metal style HPC clusters on Google Cloud Platform

Date and Time: 
April 10th @ 2:45pm
Speaker: 
Joe Schoonover

Abstract: Cloud environments have primarily focused on services for groups working with micro-services and databases. HPC ventures into cloud environments have largely been focused on conforming to containerization and micro-services. This talk focuses on a different strategy: making the cloud look more like an HPC environment. I will share progress on an open source tool that integrates with Google Cloud's deployment manager to provision a bare metal style HPC clusters with a Slurm job scheduler.

Speaker Description: 
Dr. Joseph Schoonover holds degrees in Applied Mathematics, Physics, and Geophysical Fluid Dynamics from Florida State University. His graduate studies focused on Gulf Stream separation dynamics and high order methods for computational fluid dynamics. After graduating from FSU, he held a post-doc position at the Center for Non-Linear Studies at Los Alamos National Laboratory. This is where his interest in GPU acceleration was born through mentorship activities at the Parallel Computing Summer Research Internship. After leaving CNLS and LANL, Joe became an associate scientist at CU Boulder to work at NOAA's Space Weather Prediction Center to accelerate the operational WAM-IPE code for modeling ionosphere phenomena. Joe has since moved on to a startup company, Fluid Numerics, founded with Guy Thorsby and Elizabeth Simons, that aims to help domain scientists and software developers leverage the latest computing technologies and cloud computing platforms for high performance computing.
 

Event Category:

Multi-Architecture Docker Registries: Best Practices

Date and Time: 
Friday April 12th 2019
Location: 
CG South Auditorium
Speaker: 
Julio Suarez

Speaker Description: 

Julio Suarez is an engineer in the Infrastructure Line of Business at Arm. His top priority is to work with partners to enable and improve the Arm ecosystem. This includes writing blogs, giving presentations, developing proof of concepts, and developing and running server & networking performance workloads. You will often find Julio at trade shows demoing proof of concepts of container and networking technologies running on Arm.

Event Category:

Software Testing and Testing Automation with Python

Date and Time: 
Friday April 12th 2019
Location: 
CG South Auditorium
Speaker: 
Ryan May & John Leeman

Speaker Description: 

Ryan May is a software engineer at UCAR/Unidata, working on Python software and training for the atmospheric science community. Currently, he is the core developer of the MetPy and Siphon Python packages, as well as a member of the development team for the matplotlib Python visualization library.

John Leeman is a software engineer at Unidata in Boulder, CO. He received a B.S. in meteorology, a B.S. in geophysics, and a minor in mathematics from the University of Oklahoma in 2012. While at OU he was active in gas hydrates research, and continued that work at Oak Ridge National Laboratory. Afterwards he was an intern at NASA in the GN&C Autonomous Flight Systems Branch of the Aeroscience and Flight Mechanics Division for the Morpheus lunar lander working as a programmer. John received his PhD in 2017 from Penn State in geoscience, studying earthquake physics and slow-slip.

Event Category:

Pages

Subscribe to conference-talk