Imperial Research Software Engineering (RSE) Community Meeting - May 2016

Event Information

Share this event

Date and Time

Location

Location

LT144, Huxley Building, Department of Computing

Imperial College London

180 Queen's Gate

London

SW7 2RH

United Kingdom

View Map

Event description

Description

The May 2016 meeting of the Imperial College Research Software Engineering (RSE) Community will take place on Thursday 26th May, 16:00 – 18:00 in the Huxley Building, LT144, on the South Kensington Campus. We will have a main talk from Dr Peter Vincent, Department of Aeronautics and lead of the PyFR project (http://www.pyfr.org), followed by a set of 3-minute lightning talks* and an RSE discussion.

The draft agenda is as follows:

16:00 PyFR: Next-Generation High-Order Computational Fluid Dynamics on Modern Hardware Platforms, Peter Vincent, Department of Aeronautics

16:30 3-minute lightning talks* (if you'd like to volunteer to give a lightning talk, please e-mail jeremy.cohen@imperial.ac.uk with a title and short overview of your topic)

17:00 RSE discussion

17:30 Networking; Drinks and snacks available

Main talk:

PyFR: Next-Generation High-Order Computational Fluid Dynamics on Modern Hardware Platforms

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 Graphical 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 I will present PyFR (www.pyfr.org), an open-source Python based framework for solving advection-diffusion type problems using the FR approach. The framework is designed to solve a range of governing systems on mixed unstructured grids containing various element types. It is also designed to target a range of hardware platforms via use of a custom Mako-derived domain specific language. The latest release of PyFR is able to solve the compressible Euler and Navier-Stokes equations on grids of quadrilateral and triangular elements in two dimensions, and hexahedral, tetrahedral, prismatic, and pyramidal elements in three dimensions, targeting clusters of multi-core CPUs, NVIDIA GPUs, AMD GPUs, Intel Xeon Phis, and heterogeneous mixtures thereof. Results will be presented for various benchmark and `real-world' flow problems, and scalability of PyFR will be demonstrated on clusters with 1000s of NVIDIA GPUs. Throughout the talk the importance of algorithm-software-hardware co-design, in the context of next-generation computational fluid dynamics, will be highlighted.

Project page: http://www.pyfr.org

* Lightning talks: The lightning talks are intended to provide an opportunity for you to give a brief, informal overview of a topic that you think may be of interest to other community members. You may want to introduce an application or tool you've been working on or a library that you think may be of interest to other developers. Perhaps you've recently read a great blog post that you'd like to highlight, or you have an RSE-related idea you'd like to present and get some feedback on in the susequent discussion session.

  • Lightning talks will have a limit of 3 minutes each.
  • Presenters may provide 1 slide to use as a background for their presentation.
  • If you'd like to give a lightning talk, please send a title and a couple of sentences describing your topic to jeremy.cohen@imperial.ac.uk

Date and Time

Location

LT144, Huxley Building, Department of Computing

Imperial College London

180 Queen's Gate

London

SW7 2RH

United Kingdom

View Map

Save This Event

Event Saved