A GOTO Night on Supercomputers


GOTO Night og InfinIT inviterer til gå-hjem-møde torsdag den 7. november 2013 kl. 16.30 til 19.00

A GOTO Night on Supercomputers

Supercomputers offer exceptional compute power and the move to multi-core architectures have increased this power profoundly. Just a few years ago the Fyrkat supercomputer at Aalborg University with its 6.2 teraflops, 84 nodes and total of 672 cores was considered a big machine. Today Fyrkat is still a “workhorse”, but a fairly small machine.

In June 2013, the Tianhe-2 topped the TOP500 list of fastest supercomputers in the world when it achieved 33.86 petaflops using 16,000 computer nodes with 128.000 CPU cores and almost 3.000.000 GPU cores.

At Manchester University in the UK, Professor Steve Furber is building the SpiNNaker machine with 1 million CPU cores. SpiNNaker is a novel computer architecture inspired by the working of the human brain.

All supercomputers nowadays are massively parallel systems, thus exploiting supercomputers requires additional programming skills.

To help understand the practical problems Professor Brian Vinter will look at some examples from industrial production, financial modeling and classic applications of supercomputing in science.

These talks are intended for everybody interested in the future of computing and does not require special interest or knowledge in Supercomputing

Tilmelding skal ske på GOTO Nights hjemmeside her.

Der serveres sandwich og vand midtvejs.


Talk 1:


Title:

Programming Next Generation Processors in HPC - with applications from industry, finance and science

Abstract:
The realization that processor performance is severely limited by the three walls; power-wall, frequency-wall, and memory-wall has led to the advent of a generation of processors commonly known as accelerators and more advanced conventional processors. A very common misunderstanding is that programming these next-generation processors is an exercise in parallel programming, but while parallelization is part of the solution correct memory layout and access patterns are still the primary key to performance.

In this talk we will take a look at different processors, their similarities and differences, how they may be programmed and how we may help automate the programming of them. To help understand the practical problems we will look at some examples of HPC tasks from industrial production, financial modeling and classic applications of supercomputing in science.

Bio:

Brian Vinter is Professor of Computer Science at the Niels Bohr Institute, University of Copenhagen. He is a master of computer-engineering from Aalborg University in Denmark and PhD from Tromsø¸ University, Norway. Professor Vinters professional interests are centered on High Performance Computing, with an emphasis on programmability of HPC outside conventional HPC applications.

Talk 2:


Title:

Computers and Brains

Abstract:
Computers and brains both process information, but they do so in very different ways. What are these differences, and what are the prospects for using computers to improve our understanding of brains and then using that understanding to build better computers?

Bio:

Steve Furber is considered to be the "Father-of-the-ARM processor", i.e. inventor of the chip that sits in almost all mobile phones today. His "latest" project is the SpiNNaker brain simulation machine, a machine with 1 million cores.

Steve Furber is the ICL Professor of Computer Engineering in the School of Computer Science at the University of Manchester. He received his B.A. degree in Mathematics in 1974 and his Ph.D. in Aerodynamics in 1980 from the University of Cambridge, England. From 1980 to 1990 he worked in the hardware development group within the R&D department at Acorn Computers Ltd, and was a principal designer of the BBC Microcomputer and the ARM 32-bit RISC microprocessor, both of which earned Acorn Computers a Queen's Award for Technology. Upon moving to the University of Manchester in 1990 he established the Amulet research group which has interests in asynchronous logic design and power-efficient computing, and which merged with the Parallel Architectures and Languages group in 2000 to form the Advanced Processor Technologies group.



Tid og sted

Dato:  7. november 2013
Tid:  16.30-19.00
Sted:  Aalborg Universitet, Cassiopeia, Selma Lagerlöfs Vej 300, 9220 Aalborg Ø. Lokale 0.2.11
Pris:  Gratis. Dog faktureres et no-show-gebyr på 200 kr. ved udeblivelse eller afbud på dagen.
Kontakt navn:  Tina Lindekilde Sørensen
Kontakt e-mail
Tilmeldingsfrist:  4. november 2013

Se alle Infinit arrangementer


InfinIT er finansieret af en bevilling fra Styrelsen for Forskning og Uddannelse og drives af et konsortium bestående af:
Alexandra Instituttet . BrainsBusiness . CISS . Datalogisk Institut, Københavns Universitet . DELTA . DTU Compute, Danmarks Tekniske Universitet . Institut for Datalogi, Aarhus Universitet . IT-Universitetet . Knowledge Lab, Syddansk Universitet . Væksthus Hovedstadsregionen . Aalborg Universitet