ParaFEM

The freely available portable library for parallel finite element analysis.

Read More

Wednesday, June 19, 2013

Login




Forgot login?
Register
image image image

Benchmarking ofMaterialsModelling

Engineers at EDF are developing a Materials Modelling Platform that will incorporate a number of open source software programs and provide interfaces for commercial applications.

Read the Full Story

Extreme ComputingforPDEs

Extreme engineering approaches that take into account the complex interplay of different physical processes that operate at multiple length and time scales.

Read the Full Story

Goldilocks and Dinosaur Tracks

Finite-element analysis used to investigate the extent of bias in the ichnological fossilrecord attributable to body mass.

Read the Full Story
  • 1
  • 2
  • 3
Conference
22nd June 2014 — 26th June 2014
Seville
Link to External Site
The 16th European Conference on Composite Materials, ECCM16, will be held in Seville (Spain) from the 22nd to the 26th of June 2014.
The conference is organized by the Elasticity and Strength of Materials Group of the University of Seville, under the patronage of the European Society for Composite Materials (ESCM). As you know, ECCM is Europe’s leading conference on Composite Materials and, traditionally, hosts scientists, engineers and designers, both from Academia and Industry, coming from all areas of the world.
Seville is the artistic, cultural, and administrative capital of Andalusia. It is reached by car, plane (International Airport) and high speed train, with very affordable accommodation prices. The city is more than 2,000 years old, with a big variety of fabulous monuments and an interesting history.

The important dates are as follows:

15th September 2013 - Deadline for abstract submission
28th February 2014 - Deadline for manuscripts submission
28th February 2014 - Deadline for early bird registration

Written by Lee Margetts
The 5th edition of the text book "Programming the Finite Element Method" has been completed. It is now with the Wiley production team and will be available to purchase at the beginning of the 2013/14 academic year. It is aimed at undergraduates, graduate students and practitioners in all areas that use the finite element method.

Smith, Griffiths and Margetts, "Programming the Finite Element Method", Wiley, 2013 can be preordered online.

"Programming the Finite Element Method" is the only engineering text book that introduces the reader to the basic concepts of the finite element method and then takes them on a journey where they learn how to write finite element programs to solve real world problems. Source code and libraries for around 70 example programs are provided. These are introduced step by step, building up the reader's expertise, from 1D and 2D problems that can be run on a desktop computer, to deceptively simple parallel programs that are capable of solving systems with up to a billion equations using tens of thousands of cores. These parallel programs are provided as ParaFEM.

The text book and software provide an excellent foundation for teaching and research. The software can be easily customized by the user for particular applications and interfaced with other third party tools and libraries. If you wish to explore the use of the book and the ParaFEM library (for teaching or research), please contact This e-mail address is being protected from spambots. You need JavaScript enabled to view it .


Written by Lee Margetts
Sunday 20th January 2013 is a historic date for ParaFEM. During the middle of the night, we managed to solve a problem with more than 1 billion equations, 1,023,468,720 equations to be precise!

The problem was a trivial one, a linear elastic analysis comprising a grid of 440 by 440 by 440 twenty noded hexahedral elements. However, the program did not take advantage of the regular geometry and therefore should be able to solve an arbitrarily shaped domain that is just as large.

We used HECToR to run the analysis. On 4096 MPI processes, the problem solved in less than 15 minutes. Using 8192 MPI processes, run time was 7 minutes. The compute nodes needed to be run in an "under-populated" state in order to access sufficient memory. One core on each of 4096 16-core processors was used for the 4096 process run. Two cores per processor were used in the 8192 process run.

Later in the year, we hope to report results for an arbitrary geometry and visualize the results!  
Written by Lee Margetts

Images from the Gallery