AdvOL Student Seminars and Defences
Adrian Burlacu,September 19, 2017, 15:30-16:00, ITB 201
Speaker:   Adrian Burlacu

Title:  Minimized asynchronous scalable domain transactions
AdvOL Optimization Seminars
Reza Samavi, September 19, 2017, 16:30-17:30, ITB 201
Speaker:   Reza Samavi
Department of Computing and Software
McMaster University

Title:  Optimizing data utility with privacy constraints
Fields Institute Industrial Optimization Seminar, June 6, 2017
Speakers:   Yuriy Zinchenko (University of Calgarys)
Pooyan Shirvani (TD Bank)

On the first Tuesday of each month, the Industrial Optimization Seminar is held at the Fields Institute. See the seminar series website for further information.
Sunday, 24 September 2017
Main Menu
Internal pages
Latest Theses
File Icon Computational Determination of the Largest Lattice Polytope Diameter
File Icon Computational Framework for the Generalized Berge Sorting Conjecture
File Icon Structural Factorization of Squares in Strings
Latest Reports
Visitors by region
Totals Top 20
 61 % Unknown
 14 % Commercial
 8 % networks
 6 % Canada
 3 % Germany
 2 % China
 2 % Russia
 < 1.0 % Educational
 < 1.0 % Brazil
 < 1.0 % United Kingdom
 < 1.0 % Ukraine
 < 1.0 % Poland
 < 1.0 % 
 < 1.0 % Italy
 < 1.0 % France
 < 1.0 % Netherlands
 < 1.0 % Japan
 < 1.0 % India
 < 1.0 % Australia
 < 1.0 % Organization

Visitors: 5142844
Public lecture "Petaflops, Seriously" by Dr. David Keyes
The School of Computational Engineering & Science presents a free public lecture:

Petaflops, Seriously

Dr. David Keyes
Acting Director of the Institute for Scientific Computing Research
Lawrence Livermore National Laboratory.

Monday, April 9, 2007
5:30 7:30 p.m.
McMaster University, ITB-137

Lecture poster

Sustained floating-point rates on real applications, as tracked by the Gordon Bell Prize, have increased by five orders of magnitude from 1988, when 1 Gigaflops was reported on a structural simulation, to 2005, when 100 Teraflops were reported on a molecular dynamics simulation. Various versions of Moore's Law over the same interval provide only two to three orders of magnitude of improvement for an individual processor; the remaining factor comes from concurrency, which is of order 100,000 for the BlueGene/L computer, the platform of choice for the majority of 2005 Bell Prize finalists.

As the semiconductor industry begins to slip relative to its own roadmap for silicon based logic and memory, concurrency will play an increasing role in attaining the next order of magnitude, to arrive at the long-awaited milepost of 1 Petaflops sustained on a practical application, which should occur around 2008. Simulations based on Eulerian formulations of partial differential equations can be among the first applications to take advantage of petascale capabilities, but not the way most are presently being pursued. Only weak scaling can get around the fundamental limitation expressed in Amdahl's Law and only optimal implicit formulations can get around another limitation on scaling that is an immediate consequence of Courant-Friedrichs-Lewy stability theory under weak scaling of a PDE.

Many PDE-based applications and other lattice-based applications with petascale roadmaps, such as quantum chromodynamics, will likely be forced to adopt optimal implicit solvers. However, even this narrow path to petascale simulation is made treacherous by the imperative of dynamic adaptivity, which drives us to consider algorithms and queuing policies that are less synchronous than those in common use today. Drawing on the SCaLeS report (2003-04), the latest ITRS roadmap, some back-of-the-envelope estimates, and numerical experiences with PDE-based codes on recently available platforms, we will attempt to project the pathway to Petaflop's for representative applications.
< Prev   Next >
McMaster University
McMaster University
Faculty of Engineering
Faculty of Engineering
Faculty of Science
Faculty of Science
Computing & Software
Computing & Software
Comput. Eng. & Sci.

School Website >>>

Latest Publications
Publication Downloads
Error cannot find GD extension