HiFiCoMa 2019: Workshop on Spectral/hp element method using Nektar++

We are delighted to announced that the following members of the PRISM team: Chris Cantwell, Dave Moxey,Joaquim Peiro, Spencer Sherwin were invited to organise workshop on Spectral/hp element method using Nektar++ during the symposium International Symposium on High-Fidelity Computational Methods & Applications 2019, HiFiCoMa 2019, which will be held in Shanghai. The objective of this symposium is to bring together experts in computational science and experts in engineering application to exchange new ideas and discuss development perspectives of high-fidelity methods. It also offers a platform to show the exciting scientific and engineering study undertaken in this area. The symposium will also end with a Nektar++ workshop to provide a more opportunity to understand how to apply high order spectral/hp element software.
More information on the event is available here: https://www.ishfcma.org/

Congratulations to the PRISM team members on their recent promotions

We sincerely congratulate the following colleagues on their recent promotions:

The IUTAM Symposium on Laminar-Turbulent Transition 2019

Date: 2-6 September 2019

Location: Imperial College London, UK

Scientific scope:  We have received unprecedented interest in the IUTAM Transition 2019. We received nearly 200 abstract submissions. We run
plenary sessions, individual talks and networking sessions. The scientific programme covered a range of fundamental topic areas, including:

  • Global analysis of instabilities and receptivities for complex configurations;
  • Nonlinear dynamical-systems approaches to minimal seeds and transition to turbulence;
  • Influence of multi-physics phenomena on transition: reactive flows, non-Newtonian material behaviour, interfacial flows, flows with interacting structures;
  • Novel experimental measurement and evaluation techniques for transition in complex flows;
  • Roughness-induced transition; transition from steps, gaps, junctions and other geometric imperfections;
  • Transition in hypersonic flows; prediction of thermal loads.
  • Active and passive control of flows undergoing transition; transition delay;
  • Transition mechanisms in natural and controlled environments; receptivity techniques and studies;
  • Late stages of transition and the breakdown to fully developed turbulence;
  • Transient growth problems and bypass mechanisms and their role in the transition process.

Outcomes:

  • We received very positive feedback from attendees on the well organised and attended meeting.
  • Symposium was well attended by early-career academics, post-graduate students, industry representatives, senior members of the community and invited guests.
  • 175 registered delegates of which 40% are PhDs
  • Evening reception (Monday, 2nd September) and conferenced dinner (on the 5th September) provided networking opportunities for attendees to discuss future collaborations.
  • The conference proceedings will be published by Springer.

Conference Programme:  IUTAM
Transition 2019 secured a full set of nine internationally renowned invited
plenary and keynote speakers spanning a range of topic areas:

  • Stefania Cherubini, Politechnico
    di Bari, Italy.
  • Jeffrey Crouch, The Boeing
    Company, USA.
  • Mujeeb Malik, NASA Langley
    Research Center, USA
  • Tom Mullin, Oxford University, UK
  • Helen Reed, Texas A&M
    University, USA
  • Tamer Zaki, Johns Hopkins
    University, USA
  • Andre Cavalieri, Instituto
    Tecnologico de Aeronautica, Brazil
  • Xiaolin Zhong, University of
    California, Los Angeles, USA
  • Maksim Ustinov, TsAGI Central
    Aerohydrodynamic Institute, Russia

Five of the invited speakers were from the USA and one from Brasil. The rest is from Europe.

The conference included also 67 contributing talks and 78 poster presentations. The abstract for the conference posters can be found on the following page: https://ssl.linklings.net/conferences/iutam-transition/iutamtransition2019_proceedings/views/by_sub_type.html

Nektar++ Workshop 2019

June 10th – 12th, 2019, University of Exeter, Exeter, UK.

Objective

The 4th annual Nektar++ Workshop will bring together developers and users of all experiences to hear about new and future developments in Nektar++ and the exciting science and engineering being undertaken with the code. This year, we will host the workshop at the University of Exeter.

The first two days will include a comprehensive programme of talks, which will be followed by a number of parallel informal group sessions allowing developers and users to discuss and work on specific aspects of the code.

Provisional Schedule

Talks are generally 15 minute presentations + 5 minute Q&A.

Monday 10th June

12:00 – 13:00 Registration, Welcome and Lunch

13:00 – 15:00 Talk Session 1

15:00 – 16:00 Refreshment break & Posters

16:00 – 18:00 Talk Session 2

19:00 – 21:00 Workshop dinner (self-funded)

Tuesday 11th June

08:30 – 09:30 Coffee and Arrival

09:00 – 10:20 Talk Session 3

10:20 – 10:40 Refreshment break

10:40 – 12:00 Talk Session 4

12:00 – 12:45 Lunch

13:00 – 15:00 Break-out group sessions

15:00 – 15:30 Refreshment break

15:30 – 17:30 Break-out group sessions

Wednesday 12th June

09:00 – 14:00 Break-out group sessions

Location

The workshop will be held in Streatham Court on the Streatham Campus of the University of Exeter, just outside Exeter City Centre.

Getting to campus

Directions to campus are available from the University of Exeter website. The main rail station at Exeter is Exeter St. David’s, for which there are direct services from London Paddington (around 2.5hrs) and London Waterloo (around 3.5hrs).

Getting to Streatham Court

The workshop talks will take place in Lecture Hall B of Streatham Court. Streatham Court is located at the bottom of central campus. It is building 31 on the campus map. Streatham Court is about a 20 minute walk from Exeter St. David’s train station.

Getting to The Forum

The workshop breakout sessions will take place in the Exploration Labs in the Forum Building. The Forum is located at the heart of central campus and is building 3 on the campus map. The Forum is around a 20 minute walk from Exeter St. David’s train station.

Accommodation

The University of Exeter is just outside the city centre and can be reached via a short walk of around 20-30 minutes, depending on location. The closest physical hotel is the Premiere Inn, located next to St. David’s. Other hotels include the Mercure Exeter Rougemont Hotel, the Clock Tower Hotel and the Hotel du Vin. There are also several Bed & Breakfast venues at lower costs close to the University.

Cost

The workshop is free, but accommodation and the dinner is not covered and expected to be paid by the person attending.

Registration

Registration for the workshop is now available at: https://www.eventbrite.co.uk/e/nektar-workshop-2019-tickets-57852494368

Embracing Accelerators

When: April 18th 2015
Where: Imperial College London, Skempton building, 201

Programme

10:00: Coffee

10:25: Introduction

10:30 Keynote: Alex Heinecke, Intel Parallel Computing, Santa Clara
Fighting B/F Ratios in Scientific Computing by Solving PDEs in High Order

11:30  Fabio Luporini, Imperial College London
An algorithm for the optimization of finite element integration loops

12:00  David Moxey, Imperial College London
Optimizing the performance of the spectral/hp element method with collective linear algebra operations

12:30 Lunch buffet with posters

14:00 Keynote: Karl Rupp
FEM Integration with Quadrature and Preconditioners on GPUs

15:00 Peter Vincent, Imperial College London
PyFR: An Open Source Python Framework for High-Order CFD on Heterogeneous Platforms

15:45: Discussion: Implicit vs explicit methods on accelerators.

16:15: Networking Drinks

Keynote Speakers

Alexander Heinecke
Fighting B/F Ratios in Scientific Computing by Solving PDEs in High Order

Today’s and tomorrow’s architectures follow a common trend: wider vector instructions which offer denser arithmetic intensity, but constant and therefore relatively lower bandwidth. When solving PDEs, high-order methods are a possible candidate for adopting to this hardware development. Their computing cost increases with higher order due to the higher arithmetic intensity, while relatively reducing the required memory bandwidth. Therefore, they offer an adjustable trade-off between the computational costs, required bandwidth and the accuracy delivered per degree of freedom. In this talk we examine the impact of convergence order, clock frequency, vector instruction sets, alignment and chip-level parallelism for higher order discretization on their time to solution, more precisely their time to accuracy, with respect to yesterday’s, today’s and tomorrow’s CPU architectures. From a performance perspective, especially on state-of-the-art and future architectures, the shift from a memory- to a compute-bound scheme and the need for double precision arithmetic with increasing order describes a compelling path for modern PDE solvers.

Alexander Heinecke studied Computer Science and Finance and Information Management at Technische Universität München, Germany. In 2010 and 2012, he completed internships at Intel in Munich, Germany and at Intel Labs Santa Clara, CA, USA. In 2013 he completed his Ph.D. studies at TUM and joined Intel’s Parallel Computing in Santa Clara in 2014. His core research topic is the use of multi- and many-core architectures in advanced scientific computing applications.
Awards: Alexander Heinecke was awarded the Intel Doctoral Student Honor Programme Award http://blogs.intel.com/intellabs/2012/10/25/intel-honors-9-faculty-and-25-doctoral-students-with-awards-at-intel%E2%80%99s-european-research-and-innovation-conference-eric/ in 2012. In Nov. 2012 he was part of a team which placed the Beacon System #1 on the Green500 list. In 2013 and 2014 he and his co-authors received the PRACE ISC Award for achieving peta-scale performance in the fields of molecular dynamics and seismic hazard modelling on more than 140,000 cores. In 2014, he and his co-authors were additional selected as Gordon Bell finalists for running multi-physics earthquake simulations at multi-petaflop performance on more than 1.5 million of cores.

Karl Rupp
FEM Integration with Quadrature and Preconditioners on GPUs

Efficient integration of low-order elements on a GPU has proven difficult. Former work has shown how to integrate a differential form (such as Laplace or elasticity) efficiently using algebraic simplification and exact integration. This, however, breaks down for multilinear forms or when using a coefficient. In this talk, I present results from joint work with M. Knepley and A. Terrel on how to efficiently integrate an arbitrary form using quadrature. The key is a technique we call “thread transposition” which matches the work done during evaluation at quadrature points to that done during basis coefficient evaluation. We are able to achieve more than 300GF/s for the variable-coefficient Laplacian, and provide a performance model to explain these results.
The second part of the talk discusses performance aspects of preconditioners for GPUs, in particular algebraic multigrid. While the preconditioner application maps well to the fine-grained parallelism provided by GPUs, our benchmarks indicate that GPUs have to be paired with powerful CPUs to obtain best performance.

Karl Rupp holds master’s degrees in microelectronics and in technical mathematics from the TU Wien and completed his doctoral degree on deterministic numerical solutions of the Boltzmann transport equation in 2011. During his doctoral studies, he started several interacting free open source projects, including the GPU-accelerated linear algebra library ViennaCL. After a one-year postdoctoral research position working with the PETSc-team at the Argonne National Laboratory, USA, and a research project on improving the efficiency of semiconductor device simulators at TU Wien, he is now a freelance scientist. His current activities include the GPU-acceleration of large-scale geodynamics simulations.

 

Register here.

Firedrake Workshop 7-8th June 2018

We are very happy to announce that the second Firedrake user and developer workshop will be held on 7 and 8 June 2018.

The workshop will provide the opportunity for Firedrake users and developers to engage with each other to communicate the ways that Firedrake can be used in simulation science, the latest developments in the project, and the future developments anticipated. The event will provide Firedrake users with the opportunity to spend face to face time with developers and other users.

https://www.firedrakeproject.org/firedrake_18.html

PRISM II starting 1st July 2018

We are very excited to announce that the follow-on platform grant, PRISM II, has been funded by the Engineering and Physical Sciences Research Council. It will commence on 1st July 2018.

PETSc User Meeting 4-6 June 2018

Imperial College is hosting the 2018 PETSc User Meeting. Three of the co-organisers are from the PRISM team.

PETSc, the Portable, Extensible Toolkit for Scientific Computing, is one of the world’s most widely used software libraries for high-performance computational science. We’ll hold our fourth annual user meeting (second in the EU), give tutorials on how you can leverage the functionality in PETSc for your research, and highlight science achievements made possible by advances to PETSc’s features and functionality – not only by the PETSc team but also by applications developers and designers of simulation packages that use PETSc.

https://www.mcs.anl.gov/petsc/meetings/2018/

Review on Modelling

Through his involvement with PRISM Spencer has been invited to act as an expert panel member for the Council for Science and Technology Review on Modelling Chaired by Sir Mark Walport. The review is planned to be published later this year.

Partners