Exascale Computing Challenges: Parallel-in-Time Algorithms

University of Exeter

Training in theory and implementation of parallel-in-time algorithms for academic researchers and software engineers.

Reed Hall, University of Exeter

Thank you for registering to attend, we will be in touch with further details in due course.

However if you have any questions in the meantime, please contact research-events@exeter.ac.uk.

Date and Time: Monday 9 January 2023, 11:00 - 17:00 GMTTuesday 10 January 2023, 09:00 - 17:00 GMTWednesday 11 January 2023, 09:00 - 13:00 GMTLocation: Reed Hall, University of Exeter, Streatham Drive, Exeter EX4 4QROverviewIn this workshop we will focus on time-parallel methods. Following exciting developments in both mathematical analysis and practical experience, time-parallel methods are undergoing a revival as a potentially powerful route to exploiting future massively parallel exascale supercomputers. Time-parallel methods address the question of what to do when one has reached the limits of strong scaling (decreasing wallclock time by increasing the number of processors working in parallel) through domain decomposition parallelisation in space. A key lesson from the recent literature is that the success of parallel-in-time algorithms critically depends on them being carefully adapted to the equation being solved. Much like regular timestepping methods, there are many parallel-in-time algorithms, and the right algorithm needs to be designed and selected according to the mathematical properties and applications requirements of the underlying system. This workshop will combine lectures with practical demonstrations to introduce timestepping challenges and how to overcome them using time-parallel methods, such as parareal, deferred corrections and paradiag. The practical demonstrations will be based on jupyter notebooks and some experience of using python is desirable.W﻿hat to bring?S﻿ince the course will have significant practical elements, please bring your laptop. To run code on your laptop you will need to have Python (version &gt;3) installed, along with numpy, scipy and jupyter notebook. The code will be provided via GitHub. We can support you with all of these things if necessary.S﻿cheduleMonday11.00 Introduction to the workshop: aims and practicalities, Jemma Shipton.11.30 Exascale Computing Challenges, David Acreman12.30 Lunch13.30 Why is timestepping so hard? Jemma Shipton (Lecture and practical)15.00 Coffee15.30 Introduction to parallel timestepping approaches, Jemma Shipton (Lecture)17.00 FinishTuesdayThe sessions on this day will consist of a short introduction to the method followed by some practical exercises.09.00 Deferred correction methods10.30 Coffee11.00 Paradiag12.30 Lunch13.30 Introduction to parallel profiling and benchmarking, David Acreman15.00 Break15.30 Troubleshooting and discussion session17.00 FinishWednesday09.00 Parareal10.30 Coffee11.00 How not to fool with (or be fooled by) time-parallel scaling results, Jemma Shipton11.45 Summary and discussion12.30 Close

FormatThis event is a workshop funded by the <a href="https://excalibur.ac.uk/projects/rse-training-algorithms/" target="_blank" data-msys-clicktrack="0" rel="nofollow noopener noreferrer">ExCALIBUR </a>project: RSE training in algorithms for exascale simulations.Speakers include: <ul><li><a href="https://mathematics.exeter.ac.uk/staff/js1075?sm=js1075" target="_blank" data-msys-clicktrack="0" rel="nofollow noopener noreferrer">Dr Jemma Shipton</a>, Lecturer in Mathematics (E&amp;R), University of Exeter</li><li><a href="https://www.imperial.ac.uk/people/joshua.hope-collins13" target="_blank" data-msys-clicktrack="0" rel="nofollow noopener noreferrer">Josh Hope-Collins</a>, Research Associate in Computational Mathematics, Faculty of Natural Sciences, Department of Mathematics, Imperial College London </li><li>Dr Dave Acreman, HPC Applications Analyst, Hewlett Packard Enterprise</li></ul>Organisers: <ul><li>Dr Jemma Shipton, Lecturer in Mathematics (E&amp;R), University of Exeter </li><li><a href="https://www.exeter.ac.uk/research/services/contact/staff/profile/index.php?web_id=Katie_Finch" target="_blank" data-msys-clicktrack="0" rel="nofollow noopener noreferrer">D﻿r Katie Finch</a>, Head of Research Software Engineering and Assistant Director of Research Services, University of Exeter </li></ul>

Refreshments and catering Refreshments and catering will be served throughout the event. Please do advise of any dietary requirements upon registration.

Venue Reed Hall, University of Exeter, Streatham Drive, Exeter EX4 4QR.Directions to the Streatham Campus can be found <a href="https://www.exeter.ac.uk/visit/directions/streatham/" target="_blank" data-msys-clicktrack="0" rel="nofollow noopener noreferrer">here</a>. Reed Hall is building Number 14 on the Streatham Campus map <a href="https://www.exeter.ac.uk/visit/directions/streathammap/" target="_blank" data-msys-clicktrack="0" rel="nofollow noopener noreferrer">here</a>. Travel and car parking Please note that car parking spaces are limited on campus, and we strongly encourage sustainable travel. Directions and details of travelling to campus can be found <a href="https://www.exeter.ac.uk/visit/directions/streatham/" target="_blank" data-msys-clicktrack="0" rel="nofollow noopener noreferrer">here</a>.There is dedicated car parking on site for people attending events at Reed Hall, which can be accessed from Streatham Drive. Please use postcode EX4 4QR on a Sat Nav.

AccommodationA﻿ selection of rooms are on hold at the <a href="https://all.accor.com/hotel/A0H6/index.en.shtml" target="_blank" data-msys-clicktrack="0" rel="nofollow noopener noreferrer">Mercure Exeter Rougemont Hotel </a>at the rate of £79 per a night including bed and breakfast. Please note this special rate is available until Friday 23 December. To book your accommodation, please contact the Mercure Exeter Rougemont Hotel and quote the title of the event Exascale Computing Challenges: Parallel-in-Time Algorithms via email HA0H6-RE@accor.com Or call 01392 410 237

RegistrationPlease register using the ‘register’ link on this page. If you wish to register without using Eventbrite, or if you have any questions regarding the event please contact: research-events@exeter.ac.uk.Data Protection and Privacy<ul><li>To aid organisation of this event, attendee information may be shared with relevant staff in the University e.g. a list of attendees could be provided to the University host, guest speaker.</li><li>We will store your information securely, so that we can communicate to you about the event (e.g. cancellation of event, changes to the itinerary, a satisfaction survey).</li><li>We delete our access to personal information on Eventbrite after 6 months.</li><li>More information on the University’s privacy policies can be found <a href="https://www.exeter.ac.uk/privacy/cookies/" target="_blank" data-msys-clicktrack="0" rel="nofollow noopener noreferrer">here</a>.</li><li>For more information on Eventbrite’s own privacy policy, please visit <a href="https://www.eventbrite.com/support/articles/en_US/Troubleshooting/eventbrite-privacy-policy?lg=en_US" target="_blank" data-msys-clicktrack="0" rel="nofollow noopener noreferrer">this page</a>.</li></ul>

Our {communityGuidelinesLink} describe the sort of content we prohibit on Eventbrite. If you suspect an event may be in violation, you can report it to us so we can investigate.

To report other categories of prohibited or illegal content, submit {link}

Registrations are closed

Exascale Computing Challenges: Parallel-in-Time Algorithms

More events from University of Exeter

Discover more events from University of Exeter, from Business to other experiences you might love.

Registrations are closed

Exascale Computing Challenges: Parallel-in-Time Algorithms

More events from University of Exeter

Discover more events from University of Exeter, from Business to other experiences you might love.

You might also like...

Browse more events with different dates, prices, and formats to find your next great experience.