conference-talk

Building data centric website for research projects and research groups

Date and Time: 
Thursday 2018 Apr 5th
Location: 
CG North Auditorium
Speaker: 
Amit Chourasia

Duration: Half day
Level: Beginner (No pre-requisites, minimal familiarity with command line helpful)

Speaker Description: 

Amit Chourasia is a Sr. Visualization Scientist at the San Diego Supercomputer Center, UC San Diego where he leads the Visualization Group. His work is focused on leading the research, development and application of software tools and techniques for visualization. Result and data sharing are also at a forefront of his interests, to this end he has developed a web based cloud infrastructure to enable this important and at times critical gap in scientific process via the SeedMe project.

Event Category:

Exploiting Computation and Communication Overlap in MVAPICH2 and MVAPICH2-GDR MPI Libraries

Date and Time: 
Wednesday 2018 Apr 4th
Location: 
CG Center Auditorium
Speaker: 
DK Panda

This talk will focus on the set of features available in MVAPICH2 and MVAPICH2-GDR MPI libraries to exploit overlap of computation and communication on modern clusters. Sample features will include: job start-up, point-to- point operations, RMA operations, kernel-based collectives, and non-blocking collectives (with and without core- direct support). For MVAPICH2-GDR, we will additionally focus on the use of GPU Direct RDMA, kernel-based reduction and datatype operations. Performance benefits of these features will be presented.

Speaker Description: 

DK Panda is a Professor and University Distinguished Scholar of Computer Science and Engineering at the Ohio State University. He has published over 400 papers in the area of high-end computing and networking. The MVAPICH2 (High Performance MPI and PGAS over InfiniBand, Omni-Path, iWARP and RoCE) libraries, designed and developed by his research group (http://mvapich.cse.ohio-state.edu), are currently being used by more than 2,875 organizations worldwide (in 86 countries). More than 451,000 downloads of this software have taken place from the project's site. This software is empowering several InfiniBand clusters (including the 1 st , 9 th , 12 th , 17 th , and 48 th ranked ones) in the TOP500 list. The RDMA packages for Apache Spark, Apache Hadoop and Memcached together with OSU HiBD benchmarks from his group (http://hibd.cse.ohio-state.edu) are also publicly available. These libraries are currently being used by more than 275 organizations in 34 countries. More than 25,300 downloads of these libraries have taken place. A high-performance and scalable version of the Caffe framework is available from https://hidl.cse.ohio-state.edu. Prof. Panda is an IEEE Fellow. More details about Prof. Panda are available at http://www.cse.ohio-state.edu/~panda.

Event Category:

Keynote: Designing HPC, Big Data, Deep Learning, and Cloud Middleware for Exascale Systems: Challenges and Opportunities

Date and Time: 
Tuesday 2018 Apr 3rd
Location: 
CG Auditorium
Speaker: 
DK Panda

This talk will focus on challenges and opportunities in designing HPC, Big Data, Deep Learning, and HPC Cloud middleware for Exascale systems with millions of processors and accelerators. For the HPC domain, we will discuss about the challenges in designing runtime environments for MPI+X (PGAS - OpenSHMEM/UPC/CAF/UPC++, OpenMP, and CUDA) programming models by taking into account support for multi-core systems (KNL and OpenPower), high-performance networks, GPGPUs (including GPUDirect RDMA), and energy-awareness.

Speaker Description: 

DK Panda is a Professor and University Distinguished Scholar of Computer Science and Engineering at the Ohio State University. He has published over 400 papers in the area of high-end computing and networking. The MVAPICH2 (High Performance MPI and PGAS over InfiniBand, Omni-Path, iWARP and RoCE) libraries, designed and developed by his research group (http://mvapich.cse.ohio-state.edu), are currently being used by more than 2,875 organizations worldwide (in 86 countries). More than 451,000 downloads of this software have taken place from the project's site. This software is empowering several InfiniBand clusters (including the 1 st , 9 th , 12 th , 17 th , and 48 th ranked ones) in the TOP500 list. The RDMA packages for Apache Spark, Apache Hadoop and Memcached together with OSU HiBD benchmarks from his group (http://hibd.cse.ohio-state.edu) are also publicly available. These libraries are currently being used by more than 275 organizations in 34 countries. More than 25,300 downloads of these libraries have taken place. A high-performance and scalable version of the Caffe framework is available from https://hidl.cse.ohio-state.edu. Prof. Panda is an IEEE Fellow. More details about Prof. Panda are available at http://www.cse.ohio-state.edu/~panda.

Event Category:

GISandbox: A Science Gateway for Geospatial Computing

Date and Time: 
Monday 2018 Apr 2nd
Location: 
CG Auditorium
Speaker: 
Davide Del Vento

Davide Del Vento, Eric Shook, Andrea Zonca, and Jun Wang

Speaker Description: 

Davide Del Vento is Software Engineer at National Center for Atmospheric Research, where he works since 2008 as a High Performance Computing Specialist. In this role, he provides assistance to the UCAR computing community of scientists and programmers on a variety of topics, including optimization and tuning, parallel computing, data analysis and debugging. He contributes to design, development, testing and maintenance of local software packages. Davide also serves as a Software Engineer for the XSEDE Novel and Innovative Projects (NIP) group and for the XSEDE Extended Collaborative Support Services (ECSS) program.

Moreover, Davide is the Chairman of the Software Engineering Assembly since 2010.

In a previous life, he worked as a member of the control team for the Visible InfraRed Thermal Imaging Spectrometer onboard the Venus Express mission from the European Space Agency. The mission was operational around the planet Venus from 2006 to 2014.

Event Category:

Probabilistic forecasting for weather prediction

Date and Time: 
Monday 2018 Apr 2nd
Location: 
CG Auditorium
Speaker: 
Luca Delle Monache

Probabilistic forecasts provide a distribution of the potential future states of the atmosphere. There are two main methods to generate probabilistic forecasts: one is to run multiple model realizations or multiple models with initial perturbations, and the second is to use analogs. The Analog Ensemble (AnEn) algorithm discussed in this presentation generates probabilistic predictions using a single deterministic NWP, a set of past forecast predictions, and their corresponding observations. The AnEn technique compensates for the model bias by taking past errors into account.

Speaker Description: 

Luca Delle Monache is the Deputy Director of the National Security Applications Program at NCAR. Dr. Delle Monache has research interests in applied mathematics, atmospheric chemistry, meteorology, probabilistic forecasting, applied research, and interdisciplinary research.

Event Category:

SeedMe (Stream, Encode, Explore and Disseminate My Experiments) project to support scientific data-sharing and data management

Date and Time: 
Tuesday 2018 Apr 3rd
Location: 
CG Auditorium
Speaker: 
Amit Chourasia

The need for data sharing and rapid data access has become central with the rise of collaborative research in many disciplines. For the general public, several file sharing products are available that post and share files using web browsers. But for science data and research use, these products are not well suited. While consumer products get by with manual user interfaces to add and remove a few shared files, this is not practical for sharing large numbers of science data files, like those generated during and after large-scale computation.

Speaker Description: 

Amit Chourasia is a Sr. Visualization Scientist at the San Diego Supercomputer Center, UC San Diego where he leads the Visualization Group. His work is focused on leading the research, development and application of software tools and techniques for visualization. Result and data sharing are also at a forefront of his interests, to this end he has developed a web based cloud infrastructure to enable this important and at times critical gap in scientific process via the SeedMe project.

Event Category:

MPI RMA – The State of the Art in Open MPI

Date and Time: 
Wednesday 2018 Apr 4th
Location: 
CG Auditorium
Speaker: 
Howard Pritchard Jr.

With the increasing core count on current and future processors, as well as the growing adoption of multi-level parallelism in HPC applications, realizing the full performance of RDMA capable networks presents a significant challenge to MPI implementations and associated lower level network software components.

Speaker Description: 

Howard Pritchard Jr. is Research Scientist at Los Alamos National Laboratory

Event Category:

A vision for destruction of post-Moore disruption: How I Learned to Stop Worrying and Love non-von Neumann

Date and Time: 
Tuesday 2018 Apr 3rd
Location: 
CG Auditorium
Speaker: 
Jonathan Beard

Abstrac:

Speaker Description: 

Jonathan is an experienced leader, manager, and researcher. His current research is in the area of data movement reduction hardware/software system architecture solutions for post-exascale and post-Moore systems.

Jonathan is currently a staff research engineer at ARM in Austin, Texas, advisor to FastData.io, and owner of Arkhesoft LLC. He is a graduate of Louisiana State University, The Johns Hopkins University, and Washington University in St. Louis. Jonathan holds baccalaureate degrees in Biology and International studies, a masters degree in Bioinformatics and a doctorate in Computer Science under the research direction of Dr. Roger Chamberlain.

Jonathan is a U.S. Army veteran. He served in multiple countries, in roles ranging from platoon leader and general's Aide-de-Camp to deputy director within a large successful multi-national organization. Jonathan has successfully led companies in size from 50 through 250, including managing and directing start-up operations of an organization with a multi-million USD budget.

Event Category:

Reducing dark bandwidth through data reduction near memory

Date and Time: 
Wednesday 2018 Apr 4th
Location: 
CG Auditorium
Speaker: 
Jonathan Beard

Abstract

Speaker Description: 

Jonathan is an experienced leader, manager, and researcher. His current research is in the area of data movement reduction hardware/software system architecture solutions for post-exascale and post-Moore systems.

Jonathan is currently a staff research engineer at ARM in Austin, Texas, advisor to FastData.io, and owner of Arkhesoft LLC. He is a graduate of Louisiana State University, The Johns Hopkins University, and Washington University in St. Louis. Jonathan holds baccalaureate degrees in Biology and International studies, a masters degree in Bioinformatics and a doctorate in Computer Science under the research direction of Dr. Roger Chamberlain.

Jonathan is a U.S. Army veteran. He served in multiple countries, in roles ranging from platoon leader and general's Aide-de-Camp to deputy director within a large successful multi-national organization. Jonathan has successfully led companies in size from 50 through 250, including managing and directing start-up operations of an organization with a multi-million USD budget.

Event Category:

Performance modeling results of Higher Order Symmetric Schemes

Date and Time: 
Tuesday 2018 Apr 3rd
Location: 
CG Auditorium
Speaker: 
Jack Ogaja

Performance modeling results of Higher Order Symmetric Schemes in a Limited Area model, COSMO, show that the increased stencil width associated with the schemes has insignificant impact on the computational costs when applied at climate scales. The computing time for longer time integration of the model equations is however affected by additional operations necessary for the symmetric conservative fourth order scheme.

Speaker Description: 

Jack Ogaja is a Scientific Programmer and Data Analyst at Brown University. Previously worked as a Computational Scientist at Barcelona Supercomputing Center in Spain, and Brandenburg University of Technology in Germany.

Event Category:

Pages

Subscribe to conference-talk