tag:blogger.com,1999:blog-5822805028291837738.post7035477324267614125..comments2023-12-09T03:51:33.158-05:00Comments on Various Consequences: Simulation-based Engineering ScienceJoshua Stultshttp://www.blogger.com/profile/03506970399027046387noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-5822805028291837738.post-77606128681778014202010-02-03T18:16:00.857-05:002010-02-03T18:16:00.857-05:00Slides on a presentation for a study that builds o...<a href="http://www.math.colostate.edu/~estep/education/cse/siam-sagui-glotzer.ppt" rel="nofollow">Slides on a presentation</a> for a study that builds on the NSF report cited above. <br />From slide 11:<br />Study designed to:<br />- Gather information on the worldwide status and trends of SBE&S research<br />- State of the art, regional levels of activities<br />US leadership status<br />- Opportunities for US leadership<br />- Disseminate this information to government decision makers and the research community<br />- Findings, not recommendations<br /><br />From Slide 20:<br />Top Four Major Trends in SBE&S Research<br />1. Data-intensive applications (esp Switzerland and Japan)<br /> Integration of (real-time) experimental and observational data with modeling and simulation to expedite discovery and engineering solutions<br />2. Millisecond timescales for proteins and other complex matter with molecular resolution <br />3. Science-based engineering simulations (US slight lead)<br /> Increased fidelity through inclusion of physics and chemistry<br />4. Multicore for petascale and beyond: not just faster time to solution - increased problem complexity<br /> Cheap GPUs today give up to 200x speed up on hundreds of apps!Joshua Stultshttps://www.blogger.com/profile/03506970399027046387noreply@blogger.comtag:blogger.com,1999:blog-5822805028291837738.post-37104561803475000202009-12-16T16:20:30.854-05:002009-12-16T16:20:30.854-05:00NASA tech report about squeezing out more performa...NASA tech report about squeezing out more performance from new hardware options: <a href="http://www.nas.nasa.gov/News/Techreports/2009/PDF/nas-09-003.pdf" rel="nofollow">Acceleration of a CFD Code with a GPU</a><br /><br />They showed a significant speed-up (wall-clock) with using the GPU even though they had to use a Jacobi iteration rather than an SSOR iteration, that's impressive.<br /><br />Add GPU programming to your curricula for the aspiring 'computational physics expert' (if it wasn't already there).Joshua Stultshttps://www.blogger.com/profile/03506970399027046387noreply@blogger.com