Apr 27, 2017 · 2:00 p.m.– 3:00 p.m. · Room 347, Visualization Lab
We offer an open, walk-in help session every Thursday afternoon from 2:00 - 3:00 pm in 347 Lewis Library. No appointment necessary.
For help at other times, please email email@example.com.
The Help Session is an opportunity to meet with research computing staff for one-on-one help with data visualization and programming. We can discuss visualization programs, techniques, and data formats as well as programming and cluster usage. In particular, how to effectively display your data.
If you are working with large amounts of data on the Princeton High Performance Computing environment you can learn about remote visualization from tigressdata.princeton.edu.
MAE Seminar: PyFR: High-Order Accurate Cross-Platform Petascale Computational Fluid Dynamics with Python, Friday, April 28, 3:30 PM, Maeder Hall, ACEE
Apr 28, 2017 · 3:30 p.m.– 5:00 p.m. · Maeder Hall Auditorium, Andlinger Center
Host: Luigi Martinelli, Mechanical and Aerospace Engineering
Friday, April 28, 2017 at 3:30 PM
Maeder Hall, ACEE
High-order numerical methods for unstructured grids combine the superior accuracy of high-order spectral or finite difference methods with the geometrical flexibility of low-order finite volume or finite element schemes. The Flux Reconstruction (FR) approach unifies various high-order schemes for unstructured grids within a single framework. Additionally, the FR approach exhibits a significant degree of element locality, and is thus able to run efficiently on modern many-core hardware platforms, such as graphics processing units (GPUs). The aforementioned properties of FR mean it offers a promising route to performing affordable, and hence industrially relevant, scale-resolving simulations of hitherto intractable unsteady flows within the vicinity of real-world engineering geometries.
In this talk we will present PyFR an open-source, Python-based framework for solving the Navier-Stokes equations using the FR approach at extreme scale. Results will be presented for various benchmark and "real-world" flow problems, and scalability/performance of PyFR will be demonstrated on clusters with thousands of NVIDIA GPUs. Current challenges and future directions within computational fluid dynamics, and computational mechanics in general, will also be discussed.
Freddie Witherden studied Physics with Theoretical Physics at Imperial College London between 2008–2012 earning an MSci degree with first class honors. In September of 2012 Freddie started a PhD in computational fluid dynamics in the department of Aeronautics at Imperial College London under the supervision of Dr Peter Vincent and graduated in December 2015. Early in 2016 Freddie started a postdoctoral appointment in the department of Aeronautics and Astronautics at Stanford University under the supervision of Prof. Antony Jameson. Freddie's main research interests are in the development new and novel approaches to enable the simulation of hitherto intractable flow problems at extreme scale.
Social Period outside of Maeder Hall following the seminar.
All are welcome…
2nd NSF Scientific Software Innovation Institute (S2I2) High Energy Physics / Computer Science Workshop- 5/1-5/3
May 1, 2017 · 12:00 p.m.– 1:15 p.m. · 120 Lewis Science Library
May 1-3, 2017
Lewis Science Library (Rooms 120 & 138) and Jadwin Hall (Rooms A06, 475 & 111)
Event webpage: https://indico.cern.ch/event/622920/
Next week we will be holding a workshop at Princeton which focuses on possible research collaborations between High Energy Physics and Computer Science. Many of the topics to be discussed could have elements of wider interest, however, participation by members of the Princeton community is welcome.
The meeting is free and open to Princeton graduate students, researchers/staff, faculty and its affiliates. If you intend to drop in for more than a few talks or one of the parallel sessions, please register for the event using the "Meeting Registration" form linked on the meeting webpage.
Planned discussions and sessions include:
Science Practices & Policies, Sociology and Community Issues
- Machine Learning
- Software Life Cycle / Software Engineering
- Software/Data/Workflow Preservation & Reproducibility
- Scalable Platforms
- Data Management, Access, Distribution, Organization; Data Streaming
- Data Intensive Analysis Tools and Techniques; Visualization
- Training, Education, Professional Development and Advancement
Schedule of Events: https://indico.cern.ch/event/622920/timetable/
There will be plenary talks on Monday afternoon and Wednesday morning and parallel, topical sessions on the topics above. The parallel sessions will consist of an introductory talk, short lightning talks and small group discussions.
This workshop is part of an NSF-funded project (http://s2i2-hep.org/) to conceptualize a future NSF "Sustainable Software Innovation Institute" (S2I2) which could support the software upgrades needed to meet the computational and data-intensive challenges of planned "High-Luminosity Large Hadron Collider" (HL-LHC) which will enter in operation in the 2020s. The NSF funded two such institutes (Molecular Sciences and Science Gateways) in 2016 (*).
This workshop aims to bring together a diverse set of attendees from the high energy physics (HEP) and computer science (CS) communities to understand how the two communities could work together in the context of a future NSF Software Institute aimed at supporting particle physics research over the long term. We will build on the discussions which took place at an earlier
S2I2 HEP/CS workshop (**) which took place in Dec. 2016 at NCSA/UIUC.
Questions? Email Peter.Elmer@cern.ch
May 12, 2017 · 8:30 a.m.– 5:00 p.m. · 120 Lewis Science Library
Join us on May 12th for an all-day symposium focused on emerging architectures for exascale computing as well as challenges and opportunities for applications & software in academia, national labs, and computing industries. It will bring together experts from the national labs, computing industry and academia to gain technical insights on a wide range of topics including computer architecture, systems software, languages and compilers, algorithms, and applications.
Registration is free but space is limited.
Schedule of Events
08:00 am - 08:35 am Registration and Breakfast
08:40 am - 08:45 am Welcome, James Stone and Jeroen Tromp
08:45 am - 09:30 am Amitava Bhattacharjee, Princeton University / PPPL
09:30 am - 10:15 am Victor Lee, Intel
10:15 am – 10:30 am Coffee break
10:30 am – 11:15 am Jack Wells, ORNL
11:15 am - 12:00 pm Jasmin John, NOAA
12:00 pm - 01:00 pm Lunch, 120 Lewis Science Library
01:00 pm – 01:45 pm Bill Gropp, NCSA
01:45 pm - 02:30 pm Barbara Chapman, Stony Brook University
02:30 pm - 02:45 pm Coffee break
02:45 pm – 03:30 pm Dave Turek, IBM
03:30 pm – 04:15 pm Tom Gibbs, NVIDIA
04:15 pm - 05:00 pm Jeroen Tromp, Princeton
05:00 pm - 05:30 pm Cocktail Reception, 120 Lewis Science Library
06:00 pm - 08:00 pm Dinner, Prospect House (by-invitation only)
May 17, 2017 · 4:00 p.m.– 5:00 p.m. · Room 347, Visualization Lab
NumPy is a powerful library that enables scientists to quickly build high-performance data analysis applications. This talk will cover various advanced features available in the NumPy and SciPy linear algebra APIs that will help you write more readable and efficient code, faster. Array representation, indexing, aliasing, sparse matrices, Einstein summation, and, if time permits, third-party library interoperability will be among topics covered.
Vladimir Feinberg is an undergraduate senior from the Computer Science department. Next fall, he will be attending UC Berkeley for graduate school for a PhD program in Computer Science as well.
Jun 26, 2017 · 11:00 a.m.– 7:00 p.m. · Room 347, Visualization Lab
Argonne Leadership Computing Facility (ALCF), the Blue Waters project at the National Center for Supercomputing Applications (NCSA), National Energy Research Scientific Computing Center (NERSC), Oak Ridge Leadership Computing Facility (OLCF), and the Texas Advanced Computing Center (TACC) are organizing a free, week-long "Scaling to Petascale" Institute. The goal is to prepare participants to scale simulations and data analytics programs to petascale-class computing systems in support of computational and data-enabled discovery for all fields of study.
Organizers are working to engage a large national audience. This will be accomplished by streaming the sessions to a number of collaborating organizations using full-duplex audio/video connections. Sessions will be also webcast on YouTube live, although with a lower level of support for the participants. Participants must register to attend one of the host sites or to watch the sessions on YouTube. Recordings of the presentations will be made publicly available after the institute is completed.
Jun 28, 2017 · 10:00 a.m.– 3:30 p.m. · Princeton Center for Theoretical Science, 407 Jadwin Hall
This two-day workshop is intended to provide a practical introduction to the broad topic of parallel computing for scientific/numerical codes. Both MPI and OpenMP, two of the most popular tools for compiled-language parallel programming will be covered. The primary focus of the workshop will be features and topics most useful for beginning parallel programmers.
The first day will cover MPI, including environment management, point-to-point communication, and collective communication routines. The second day will cover OpenMP constructs for specifying parallel regions, work sharing, synchronization, and environment management. Covered throughout the workshop will be general strategies for writing and scaling parallel code as well as tips for optimizing and debugging. Examples will be provided in C and C++ and multiple lab exercises will allow participants to hands-on experience designing, writing, compiling, and executing parallel code on a Princeton Research Computing High Performance Computing (HPC) Cluster.
Level/Prerequisites: This workshop assumes no prior experience with parallel programming but does require some basic familiarity with C or C++ and knowledge of basic programming concepts. Because the Fortran MPI and OpenMP implementations are very similar to C/C++, Fortran programmers are encouraged to attend (though a quick overview of C syntax would be recommended). Some experience working in a Linux command line environment is likely necessary for the hands-on sections. Registrants must have a Princeton University netID, as this is required to gain access to the HPC cluster. Participants should bring their own computers with an ssh client already installed.
Because of hands-on nature of this workshop please bring:
1. A laptop with a wireless connection and an ssh client already installed. Linux and mac should already have one. Windows users will need a client, e.g. putty, cygwin, etc.
2. An account on adroit. If you do not already have one you can register here: https://www.princeton.edu/researchcomputing/computational-hardware/adroit/registration. In the “specify reason for use” write “Intro to Parallel Computing Workshop”. This is not automated so please do not wait until the last minute to register!
Ian Cosden is a Manager, HPC Software Engineering and Performance Tuning,in the Research Computing department at Princeton University. Prior to his current position he was a Research Computing Software & Programming Analyst where he worked with researchers to help build, develop, debug, and optimize serial/parallel scientific codes. He has a Ph.D. in Mechanical Engineering from the University of Pennsylvania where he developed the first highly-parallel hybrid atomistic-continuum model for liquid-vapor phase change.
Please register online at the training website, www.princeton.edu/training or contact Andrea Rubinstein at firstname.lastname@example.org /258-1397.