Python increasingly is being utilized as a powerful scientific
processing language. It successfully has been used as a glue language
to drive simulations written in C, Fortran or the array
manipulation language provided by the NumPy package. Originally
Python only was used as a glue language because the original Python
implementation was relatively slow. With the recent progress in the
PyPy project that is showing significant performance
improvements in each release, Python is nearing performance comparable
Speaker Description:
Maciej Fijałkowski is a core PyPy developer, leading the implementation of NumPy on PyPy. He has been contributing to the PyPy project since 2005, working on multiple areas, including Just In Time compiler, assembler generation, garbage collector and more. He also has experience working on the SKA (Square Kilometer Array) project in South Africa, which aims to be the biggest radio telescope ever built.
Raffaele Montuoro, Ping Chang, R. Saravanan, L. Ruby Leung
Speaker Description:
Dr. Raffaele Montuoro is Senior Lead IT Consultant with the Texas A&M University Supercomputing Facility in College Station, TX. He joined Texas A&M University in 2004, after working as IT consultant for Eutelsat SA in Paris, France. Dr. Montuoro holds a PhD in Theoretical Chemistry from the Scuola Normale Superiore di Pisa, Italy, and has developed innovative numerical models used for accurate calculations of photoionization phenomena. In 2010, some of his recent work in code optimization has been featured in the national press. Dr. Montuoro is currently collaborating with investigators at Texas A&M and PNNL to create a comprehensive high-resolution coupled regional climate model for simulations over the Atlantic Ocean.
Zbigniew Piotrowski is a Postdoc in the Institute for Mathematics Applied at the National Center for Atmospheric Research (NCAR), in Boulder, Colorado. Before arriving to NCAR in 2010 he held an appointment at the Institute of Meteorology and Water Management, Poland. His current interests include petascale simulations of geophysical flows of all scales, high resolution anelastic models for numerical weather prediction, and numerical realizability of thermal convection.
Service Oriented Architectures (SOA) are increasingly being used by a number of groups throughout UCAR for integration between systems using web services like SOAP and REST. This talk will provide a brief overview and demo of several examples of SOA solutions in use at UCAR to give engineers a sense of the growing number of options available to them to access institution-wide data and services. It will also highlight areas ripe for further application of SOA approaches.
Speaker Description:
Markus Stobbs is Group Head of the Web Engineering Group (WEG) at NCAR where he has worked since 2002 hosting websites for the institution and building custom web applications. He also manages supercomputer accounting for CISL. He has been building websites since 1993, the very early days of the Web. Previously, he was Senior Webmaster for Incyte Genomics where he built web apps for genomic scientists, and before that was Webmaster at Claris, an Apple Computer subsidiary where he managed a number of online support forums and built Claris' first public and Intranet websites. Markus is an avid hiker, mountain biker, photographer, and poet.
Designing, porting and optimizing applications for rapidly evolving computing systems
is often complex, ad-hoc, repetitive, costly and error prone process due to an enormous
number of available design and optimization choices combined with the complex interactions
between all components.
I will present a possible solution to this fundamental problem based on collective
participation of users combined with empirical tuning and predictive modeling.
I will describe public Collective Tuning Repository and new projects for collaborative
Speaker Description:
Grigori Fursin obtained BS in electronics and MS in computer engineering from MIPT (Russia), PhD in computer science from the University of Edinburgh (UK), and is currently a tenured research scientist at INRIA (France). In 2010-2011, Grigori was on sabbatical helping to establish Intel Exascale Lab in France and serving as the head of application characterization and optimization group.
Grigori's main interests are in code and architecture characterization and auto-tuning for performance, power and other characteristics using empirical, statistical, collective and machine learning techniques. He was the technical leader of the MILEPOST project (2006-2009) developing the first public machine learning based compiler. Grigori also established collaborative portal (cTuning.org) with the repository and open source tools to share and systematize knowledge about program and architecture design and optimization. Grigori is collaborating with multiple companies and universities including Intel, CAPS Entreprise, IBM, Google, University of Edinburgh, UIUC, Paris South University, ICT and others.
Paul Madden is a software engineer at NOAA's Earth System Research lab, working through the Cooperative Institute for Research in Environmental Sciences at the University of Colorado, primarily in support of numerical-weather prediction modeling efforts. I have been working in software development for the past four years, after having spent fifteen years in system administration. Paul is also a masters student in CU's Department of Computer Science, and expect to graduate this spring.
Learning the syntax of a new language is easy, but learning to think under a different paradigm is hard. This session helps you transition from a Java writing imperative programmer to a functional programmer, using Java, Clojure and Scala for examples. This session takes common topics from imperative languages and looks at alternative ways of solving those problems in functional languages. As a Java developer, you know how to achieve code-reuse via mechanisms like inheritance and polymorphism.
Speaker Description:
Neal Ford is Software Architect and Meme Wrangler at ThoughtWorks, a global IT consultancy with an exclusive focus on end-to-end software development and delivery. He is also the designer and developer of applications, instructional materials, magazine articles, courseware, video/DVD presentations, and author and/or editor of 6 books spanning a variety of technologies, including the most recent The Productive Programmer. He focuses on designing and building of large-scale enterprise applications. He is also an internationally acclaimed speaker, speaking at over 250 developer conferences worldwide, delivering more than 1000 talks. Check out his web site at nealford.com. He welcomes feedback and can be reached at nford@thoughtworks.com.
Dan Nagle will discuss coarrays as defined by the Fortran 2008 standard.
Coarrays is a PGAS scheme for scalable parallel computation.
I will have some example codes in OpenMP, MPI and coarrays
for comparison. I will review the wish-list for the future enhanced
coarrays technical specification which is expected to be published
within the next two years.
Speaker Description:
Daniel Nagle is the chair of PL22.3 (formerly J3) Fortran Standard
Committee. He got his PhD in Computational Science from GMU and is
working in UCAR’s Consulting Service Group now. He has been using and
teaching Fortran since the '60s and has been parallel programming in
Fortran and other languages since the '80s
Robert Jacob is a Computational Climate Scientist in the Mathematics and Computer Science division and a fellow in the Argonne-University of Chicago Computation Institute. He received his Ph.D. in Atmospheric Science from the University of Wisconsin-Madison. Throughout his career, he has been strongly involved in the development and application of climate models. He is the lead developer of the Fast Ocean Atmosphere Model, a fully coupled climate model, and a co-developer of the Model Coupling Toolkit and the CESM coupler.
Project Management in High Performance Computing (HPC) and specifically Research and Development (R&D) projects have traditionally been executed leveraging monolithic waterfall methodologies or sometimes spiral methodologies. While this has been the de-facto standard for decades with mixed results, Agile should be considered as a viable option for large-scale R&D HPC projects as well. Using the Agile methodology for projects with so many unknowns is as good an application of the methodology as the traditional application and for largely similar reasons.
Speaker Description:
Jessica A. Popp, PMP
Head of Project Management Office, Whamcloud, a company doing development for the Lustre filesystem http://www.whamcloud.com
