conference-talk

The Analog Ensemble (AnEn) technique for probabilistic forecasts

Date and Time: 
Thursday 2018 Apr 5th @ 8:30am OPTIONAL system set up
Thursday 2018 Apr 5th @ 9am until 12 noon tutorial session
Location: 
CG Auditorium
Speaker: 
Laura Clement-Harding

Laura Clemente-Harding, Luca Delle Monache, Guido Cervone, Martina Calovi, Weiming Hu, Mehdi Shahriari.

Speaker Description: 

Laura Clemente-Harding is a graduate student at the Pennsylvania State University (PSU) and a research scientist the Geospatial Research Laboratory.

Luca Delle Monache is the Deputy Director of the National Security Applications Program at NCAR.

Guido Cervone is an Associate Professor and Associate Director for the Institute for CyberScience at PSU.

Martina Calovi is a Postdoctoral Scientist at PSU.

Weiming Hu is a graduate student at PSU.

Mehdi Shahriari recently graduated from PSU.

Event Category:

Decoupled MapReduce using MPI one-sided communication

Date and Time: 
Tuesday 2018 Apr 3rd
Location: 
CG Auditorium
Speaker: 
Sergio Rivas-Gomez

In this talk, we present an on-going work that proposes the integration of MPI one-sided communication in MapReduce frameworks. By using a decoupled strategy, we aim to overlap computations with communications during the Map and Reduce phases. Hence, we effectively increase the performance in situations where the workload is unexpectedly unbalanced. In addition, we present a novel mechanism for decentralized job-stealing, which enables MPI processes to obtain ownership of input datasets.

Speaker Description: 

Sergio Rivas-Gomez is a Ph.D. student in High-Performance Computing at KTH Royal Institute of Technology (Sweden) under the supervision of Erwin Laure and Stefano Markidis. His research focuses on the integration of unified programming interfaces that enable new mechanisms for handling high-performance I/O on Exascale. Sergio is also part of the SAGE Project, one of the leading Exascale efforts of the European Union.

Event Category:

Reproducible Petascale Deep Learning Workflows Reproducibility is a feature of all good science

Date and Time: 
Tuesday 2018 Apr 3rd
Location: 
CG Auditorium
Speaker: 
Benjamin Liebersohn

At Oak Ridge National Lab (ORNL), leadership computing has included container usage in high performance computing, (HPC) with the intent to develop reproducible HPC applications and workflows. The Geographic Information Sciences and Technology (GIST) group at ORNL has been using containers in its HPC deep learning workflows, allowing its researchers to quickly create and adapt computing environments as their needs evolve.

Speaker Description: 

Benjamin Liebersohn is Researcher at Oak Ridge National Laboratory

Event Category:

Cloud-Hosted Real-time Data Services for the Geosciences (CHORDS)

Date and Time: 
Tuesday 2018 Apr 3rd
Location: 
CG Auditorium
Speaker: 
Aaron Botnick

CHORDS, an EarthCube Building Block, addresses the ever-increasing importance of real-time scientific data, particularly in mission critical scenarios, where informed decisions must be made rapidly. Many of the phenomenon occurring within the geosciences, ranging from hurricanes and severe weather, to earthquakes, tsunamis, volcanoes and floods, can benefit from better handling of real-time data.

Speaker Description: 

Aaron Botnick is a software engineer at NCAR's Earth Observing Laboratory, responsible for development and support of the CHORDS application. Aaron has previously worked in various software engineering roles developing both web and desktop applications.

Event Category:

Open Source Tools for Point Cloud Processing, Storage, Subsetting, and Visualization

Date and Time: 
Tuesday 2018 Apr 3rd
Location: 
CG Auditorium
Speaker: 
Kevin Beam

In this talk we provide an overview of a set of powerful open source tools available to process, store, subset, and visualize large volumes of point cloud data obtained through LIDAR instrumentation. We discuss Valkyrie, an experimental service developed at NSIDC, which can be used to subset and serve LIDAR point-cloud data obtained from NASA's Operation IceBridge flights that take place over the Arctic and Antarctic each year.

Speaker Description: 

Kevin Beam is a Professional Research Assistant / Software Engineer at the National Snow and Ice Data Center (NSIDC) at the University of Colorado Boulder. He has an interest in all things Python, JavaScript, and Linux

Event Category:

Bursts and Cascades: Scaling Up Scientific Data Analysis

Date and Time: 
Tuesday 2018 Apr 3rd
Location: 
CG Auditorium
Speaker: 
Seth McGinnis

When scientists are working on an analysis or a data transformation, they often work with a single test case or dataset at a time, and scale up to multiple cases by simply running their code multiple times. This approach works well up to a point, but new approaches are necessary in the realm of Big Data. Many data analysis tasks are embarrasingly parallel and well-suited to scaling up via parallelism.

Speaker Description: 

Seth McGinnis is an Associate Scientist IV with joint appointments in CISL and RAL. As the Data Manager for the NARCCAP and NA-CORDEX data collections, he makes the output from regional climate models usable by and available to people who need information about climate change in North America. His research focuses on bias correction, interpolation, data access, and other issues affecting the practical use of model output by non-specialists.

Event Category:

Analyzing Large Radar Datasets Using Python

Date and Time: 
Monday 2018 Apr 2nd
Location: 
CG Auditorium
Speaker: 
Robert Jackson

Robert Jackson, Scott Collis, Zach Sherman, Giri Palanisamy, Scott Giangrande, Jitendra Kumar, Joseph Hardin

Speaker Description: 

Bobby Jackson is a postdoctoral researcher at Argonne National Laboratory. He received his B.S. in Math and Computer Science and Ph.D. in Atmospheric Sciences at the University of Illinois at Urbana-Champaign. He now specializes in developing algorithms to analyze large scanning radar datasets using HPC platforms.

Event Category:

Hyperspectral Image Analysis with a Functional Data Model

Date and Time: 
Monday 2018 Apr 2nd
Location: 
CG Auditorium
Speaker: 
Doug Lindholm

Hyperspectral imaging instruments are producing large volumes of data that pose a challenge for data analysis. Hylatis is a NASA funded project to build interactive hyperspectral image analysis tools backed by cloud resources. Hylatis is built on the LaTiS software package which implements a Functional Data Model (FDM) that builds on the relational data model and relational algebra used by many database systems. The FDM captures the functional relationship between independent and dependent variables that are inherent in scientific data.

Speaker Description: 

Doug Lindholm has been wrangling scientific data at CU or UCAR since the early 1990s. Years of scientific programming and data management have contributed to an appreciation of software engineering and useful data abstractions. Doug currently works on the Web Team at CU's Laboratory for Atmospheric and Space Physics developing data services.

Event Category:

Functional Programming Concepts for Data Processing

Date and Time: 
Monday 2018 Apr 2nd
Location: 
CG Auditorium
Speaker: 
Chris Lindholm

Functional programming concepts and languages are becoming increasingly popular in industry, because they successfully address challenges related to complexity, maintainability, and correctness of software, especially as more applications move to parallel and distributed environments. This talk will introduce basic concepts important to functional programming, such as immutability and referential transparency, and how they might be applied to data processing or other scientific applications.

Speaker Description: 

Chris Lindholm has worked in the Data Systems division of CU's Laboratory for Atmospheric and Space Physics (LASP) for eight years. During that time, he has worked on data processing systems and data analysis for three NASA missions. He currently works on data access infrastructure using Scala (a functional programming language) and working out ways to use functional programming to improve scientific software.

Event Category:

Visualization of complex spacecraft orbits and multi-instrument observations

Date and Time: 
Monday 2018 Apr 2nd
Location: 
CG Auditorium
Speaker: 
Alexandria DeWole

Complex spacecraft orbits and multi-instrument observations can be challenging to visualize with traditional 2D plots. To facilitate the exploration of planetary science data, we have developed a set of web-based interactive 3D visualizations for the MAVEN and MMS missions using the free CesiumJS library. The Mars Atmospheric and Volatile Evolution (MAVEN) mission has been collecting data at Mars since September 2014. The MAVEN3D project allows playback of one day’s orbit at a time, displaying the spacecraft’s position and orientation.

Speaker Description: 

Alex DeWolfe manages the Science Data Centers for NASA’s MAVEN mission to Mars and the Emirates Mars Mission. She coordinates the design, development, implementation, and operation of science data processing systems for missions and instrument teams. She has a particular interest in scientific data visualization and oversees several projects related to developing visualization software.

Event Category:

Pages

Subscribe to conference-talk