FL2-1001 Small Seminar (next to the entrance of the building)
Speaker:
Beau Paisley
Developing an HPC application can be a challenging task - especially when it comes to fixing bugs, optimizing workload or even resolving both type of issues simultaneously. Those challenges are made easier with Allinea Tools. Using our environment, it is now possible for developers to adopt instantly efficient and scalable development tools and to focus immediately on their core activity - which can be science, benchmarks, support or even more. During this tutorial, we will review the capabilities of Allinea Tools and how to use them on yellowstone.
Speaker Description:
Beau is a computer science and mathematics graduate from the College of William and Mary and performed graduate studies in Electrical Engineering at Purdue University. Beau has over twenty five years of experience in development, marketing, and sales roles with research, academic, and startup organizations. He has previously held positions with NCAR, Applied Physics Lab, and several startup and early growth technical computing companies. Beau is now a Support Engineer with Allinea Software.
David Brown, Wei Huang, Mary Haley and Rick Brownrigg
PyNIO is a Python package that allows read and/or write access to a variety of scientific data formats, including NetCDF 3/4, HDF 4/5, HDFEOS 2/5, GRIB 1/2, and shapefiles. It provides a consistent NetCDF-like data model for all formats and through its interface to the NumPy array module allows for easy access to the power of Python scientific computing tools. In combination with mpi4py it has provided significant speedups for a number of "big data" processing tasks involving climate model data. Significant development to modernize the code is ongoing.
Py.test is a full feature tool to test your Python code. It offers a simple way to get started writing test in your work. Py.test scales from simple unit testing to complex functional testing. It is already widely used in many small and large projects. This tutorial will be a short introduction to general testing terminology and py.test tool. It will cover a basic py.test usage, starting from unit testing and the assert statement. Parametrization of arguments for a test function will be presented as a way to test a wide range of arguments’ values at the same time.
Speaker Description:
Dorota Jarecka is an Assistant Professor at the University of Warsaw (Warsaw, Poland) and a Visitor within the Mesoscale and Microscale Meteorology Division of NCAR. She received her PhD in Physics from University of Warsaw in September 2012. Her research interests include cloud microphysics, atmospheric numerical simulations and scientific computing with Python. She just started a new project dedicated to develop and test microphysical schemes in numerical models.
Packaging and distribution is essential to the success of open-source Python projects. This tutorial covers the process of turning a simple Python script into a package that can be uploaded to the Python Package Index. In addition to basic and advanced use of distutils and Setuptools, it explores integration of documentation, unit testing, code quality assurance, continuous integration, and Python 2/3 compatibility. By the end of the tutorial, participants will have a full Python package which can be used as a basis for future Python projects.
Speaker Description:
Sean Fisk is graduate student at Grand Valley State University studying for his Master’s degree in Computer Science. In summer 2013, he was an NCAR intern in the SIParCS program of Computational and Information Systems Laboratory.
Computer models and remote sensing missions often generate datasets that are too large to store and analyze on desktop computers or even on local compute clusters. A traditional approach to exploring such datasets is to run the analysis program on a remote machine and display the output on the desktop computer via X windows. The tunneling option of the secure shell connection can be used to facilitate remote display of graphics generated by analysis software like Matlab, IDL, NCL etc.
Speaker Description:
Saravanan has been a professor in the Department of Atmospheric Sciences at Texas A&M University since 2005. Previously, he worked as a scientist in CGD at NCAR. He received his Ph.D. in Atmospheric and Oceanic Sciences from Princeton University in 1990. His research involves the use of supercomputers for numeric al modeling and data analysis to study past, present, and future climates. He also dabbles in open source software and teaches courses in meteorology, climate, and introductory programming (using Python)
The aim of this tutorial is to present an overview of the most useful and important techniques and tools that can be used to efficiently work with Python in HPC. Different tips and approaches that developers can implement to improve the performance of their Python codes will be provided, as well as information on how to take advantage of multicore and distributed programming. This tutorial will also focus on different alternatives for efficiently developing, profiling and running Python codes on Stampede, however, most of the topics covered will be applicable to any other machine.
Speaker Description:
Antonio joined TACC in 2014 as a Research Associate in the High Performance Group. Previously, he was Postdoctoral Fellow at CSIRO (Commonwealth Scientific and Industrial Research Organization) at the Mathematics, Informatics, and Statistics Division, now Computational Informatics, in the Operations Research Group in Clayton, Victoria, Australia. He obtained his PhD in Computer Science in 2011 from the University of Extremadura, Spain, in collaboration with the National Fusion Laboratory, where he worked as Research Project Officer.
Johnny Lin graduated from Stanford University with a B.S. in Mechanical Engineering and an M.S. in Civil Engineering-Water Resources. After working as an environmental engineer, he returned to school and received his Ph.D. in Atmospheric Sciences from UCLA, as a student of David Neelin. His atmospheric science research is focused on stochastic convective parameterizations, ice-atmosphere interactions in the Arctic, and simple frameworks for modularizing climate models.
He is also working on a book on environmental ethics and helps coordinate the PyAOS mailing list and blog (pyaos.johnny-lin.com), an effort at building up the atmospheric and oceanic sciences Python community. Currently he is Senior Lecturer and Director of Undergraduate Computing Education at University of Washington, Bothell
FL2-1001 Small Seminar (next to the entrance of the building)
Speaker:
John Linford
The TAU Performance System is a powerful and highly versatile profiling and tracing tool ecosystem for performance analysis of parallel programs at all scales. Developed for almost two decades, TAU has evolved with each new generation of HPC systems and presently scales efficiently to hundreds of thousands of cores on the largest machines in the world. TAU has helped many projects scale up successfully on systems at Oak Ridge Leadership Computing Facility (OLCF), the National Energy Research Scientific Computing Center (NERSC), the Argonne Leadership Computing Facility (ALCF), and others.
Speaker Description:
Dr. John Linford is a Scientist at ParaTools, Inc. He received his Ph.D. from Virginia Tech, where his dissertation on accelerating atmospheric modeling through emerging multi-core technologies was selected as the outstanding doctoral dissertation of 2010. John has developed a meta-programmer for chemical kinetic simulation, airborne signal processing applications, rotocraft engineering tools, and toolkits for porting parallel HPC applications to cloud computing platforms. John helps develop the TAU Performance System and has contributed to the Scalasca project and the MoinMoin project.
Python works great as a traditional procedural language, and in that mode, can do all the atmospheric and oceanic sciences work we desire. But if we use just a few more of its features---dynamic data structures like dictionaries and how any object (even functions) have the same status---we can make our models more flexible and powerful and create modeling experiments that would otherwise be difficult to do.
Johnny Lin graduated from Stanford University with a B.S. in Mechanical Engineering and an M.S. in Civil Engineering-Water Resources. After working as an environmental engineer, he returned to school and received his Ph.D. in Atmospheric Sciences from UCLA, as a student of David Neelin. His atmospheric science research is focused on stochastic convective parameterizations, ice-atmosphere interactions in the Arctic, and simple frameworks for modularizing climate models.
He is also working on a book on environmental ethics and helps coordinate the PyAOS mailing list and blog (pyaos.johnny-lin.com), an effort at building up the atmospheric and oceanic sciences Python community. Currently he is Senior Lecturer and Director of Undergraduate Computing Education at University of Washington, Bothell
Python works great as a traditional procedural language, and in that mode, can do all the atmospheric and oceanic sciences work we desire. But if we use just a few more of its features---dynamic data structures like dictionaries and how any object (even functions) have the same status---we can make our data analysis code substantially more flexible and powerful.
Speaker Description:
Johnny Lin graduated from Stanford University with a B.S. in Mechanical Engineering and an M.S. in Civil Engineering-Water Resources. After working as an environmental engineer, he returned to school and received his Ph.D. in Atmospheric Sciences from UCLA, as a student of David Neelin. His atmospheric science research is focused on stochastic convective parameterizations, ice-atmosphere interactions in the Arctic, and simple frameworks for modularizing climate models.
He is also working on a book on environmental ethics and helps coordinate the PyAOS mailing list and blog (pyaos.johnny-lin.com), an effort at building up the atmospheric and oceanic sciences Python community. Currently he is Senior Lecturer and Director of Undergraduate Computing Education at University of Washington, Bothell