Extreme Engineering - Opportunities Using Petaflop Computing
17th July 2008, Tower Seminar Room at STFC Daresbury Laboratory
(Organiser: Prof Dave Emerson)
This meeting will focus on the challenges and opportunites offered by petaflop computing. To achieve this level of capability will require a substantial effort to ensure codes can scale to 100,000 processors and beyond. Several countries are already at an advanced stage in the deployment of such capability, particularly the USA where the first linpack petaflop benchmark was achieved on June 9th and has entered the number 1 spot in the Top 500. Japan are also making important developments in deploying this advanced level of capability. Europe has also been involved in activities for deploying petascale computing, such as the Partnership for Advanced Computing in Europe (PRACE) which is a project funded in part by the EU's 7th Framework Programme to look at a European HPC infrastructure.
This meeting is being organised by Collaborative Computational Project 12 (CCP12) and is aimed at raising awareness of the potential benefits of petaflop computing and to develop an ambitious vision for the near term (~5 years), which will take us to the end of the HECToR service, and the long term, through to 2020 when the exaflop era will be approaching. The primary aim of this meeting is to develop this vision and a strategy that will enable computational engineering applications codes to scale far beyond currently available technology. To achieve this we need to identify what the current capabilities are and where gaps exist. In particular:
- what theoretical developments are needed?
- what hardware will deliver the scientific breakthroughs?
- what are the key codes and functionality - where are the gaps?
- what needs to be done to achieve a critical level of performance?
To maximise the long term aims of the meeting, attendance is initially by invitation only. If you have an interest in attending the meeting or would like to know more details, please contact Mrs. Shirley Miller (S.Miller@dl.ac.uk).
Speakers will include...
Dr. Mike Ashworth (STFC Daresbury Laboratory)
Prof. Ken Badcock (Liverpool University)
Prof. David Emerson (STFC Daresbury University)
Prof. Neil Sandham (University of Southampton)
Dr. Jamil Appa (BAE Systems)
Prof. Robert M. Kirby (School of Computing, University of Utah)
The outline programme is given below. The meeting is to encourage discussion and establish what the current state of the art is, rather than present progress with personal research projects. The goal is to look ahead and see what is beyond the horizon and what we need in place to tackle tomorrow's problems. Natuarally, this should be backed by what has been done previously to ensure future goals and objectives are ambitious but also based on reality. The vision for 2020 needs to also factor in the prospect of exascale computing and what impact that will have.
10:00 - 10:20: Dr M.Ashworth (STFC)
10:20 - 10:40: Dr R.Blake (STFC)
10:40 - 11:00: Prof D.Emerson (STFC)
11:00 - 11:30: BREAK and DISCUSSION
11:30 - 11:50: Prof K.Badcock (Liverpool University)
11:50 - 12:10: Dr J.Appa (BAe Systems)
12:10 - 12:30: Prof N.Sandham (Southampton University)
12:30 - 13:30: LUNCH and DISCUSSION
13:30 - 13:55: Prof M.Kirby (University of Utah)
13:55 - 14:20: Dr Y.Fornier (EDF)
14:20 - 14:30: Dr K.Duraisamy (Glasgow University)
14:30 - 14:40: Prof P.Stansby (Manchester University)
14.40 - 14:50: Prof P.Gaskell (Leeds University)
14:50 - 16:00: OPEN FORUM and DISCUSSION
All enquiries about this workshop should be directed to...
Prof. David Emerson
STFC Daresbury Laboratory
Daresbury Science & Innovation Campus
Warrington WA4 4AD
Tel. +44 (0)1925 603221
Fax. +44 (0)1925 603634