David E. Goldberg and Kumara Sastry are presenting the results about the results done on the billion bit or burst initiative. In they own words:

The push for better understanding and design of complex systems requires the solution of challenging optimization problems with large numbers of decision variables. This note presents principled results demonstrating the scalable solution of a difficult test function on instances over a billion variables using a parallel implementation of a genetic algorithm (GA). The problem addressed is a noisy, blind problem over a vector of binary decision variables. Noise is added equaling a tenth of the deterministic objective function variance of the problem, thereby making it difficult for simple hillclimbers to find the optimal solution. The genetic algorithm used - the compact GA - is able to find the optimum in the presence of noise quickly, reliably, and accurately, and the solution scalability follows known convergence theories. These results on noisy problem together with other results on problems involving varying modularity, hierarchy, and overlap foreshadow routine solution of billion-variable problems across the landscape of complexity science.

Filed by admin at 3:22 pm under Events
No Comments

Ray Plante is covering the large synoptic survey telescope project. Lost of possibilities opening from its usage. More questions about the origins of the space…

Filed by admin at 2:21 pm under Events
No Comments

Donna Cox is firing a very interesting overview of the methodology art and science her group has been using at NCSA. As a study case she presented the F3 tornado in South Dakota they have been working. It is a clear compelling case for visualization as a powerful tool for science and research to gain insight on the problems at hand. And the same developed techniques were also applied to oceanographic data. She continued presenting some of the work they have done on astronomy (very fascinating animations). A key concept: visual idiom.

Filed by admin at 2:15 pm under Events
No Comments

David Barkai is introducing the Intel view about the multicore era already started. He is kicking off reviewing what is coming from Intel. They are currently in 65nm, then 45nm, and 32nm. On the way down they compact the X86 and slowly introducing a new architecture. The change to 45nm also implies the introduction of a new transistor transistor slashing the amount of power required by half. Also, details of the new quad cores lining out in the new 45nm platform. The second half of next year, Intel will also release a new architecture. The numbers: from 1 to 16+ cores per chip. Intel is also involved in working on accelerators for PCIX. Very interesting glance to the future coming. Very interesting challenge to figure out what kind of implementations we should focus on.

Filed by admin at 11:59 am under Events
No Comments

NCSA deputy director is reviewing the center history about HPC, and the need to bring people in before the technology is out. Key people + innovative hardware = road to success.

Filed by admin at 11:09 am under Events
No Comments

The second PSP speaker is Dan Fay, more on the technical computing efforts by Microsoft. His focus is to provide and end-to-end solution from solution creation to the end publication (the focus here is on science on research). A claim, provide access to archived data sets and workflows that process them. Also, how do you collaborate with scatter communities, and how their datasets and analytics can be plugged into the overall picture. One project is moving the business intelligence from SQL server in to research intelligence. Basically, standardizing the data storage and recovering to allow analytics to go over it.

Filed by admin at 10:04 am under Events
No Comments

Kyril is going over the drop of the cost of clusters and how that is penetrating companies. Most of the big solvers (e.g. Fluent) are being integrated into the user experience, being the user the one starting the HPC computation. Another key point he mentioned is the raising of workflows as the glue for enterprise and HPC approaches. Then he went over the products Microsoft is pushing for HPC.

Filed by admin at 9:16 am under Events
No Comments

Yesterday NCSA 2007 PSP meeting started. DITA was there demoing the recent advances on the DISCUS project. We cover some of our latest work on tracking the blogosphere. Lots of interesting discussions with PSP partners from Intel and Dere among others. Today, very interesting presentations are line up.

Filed by admin at 9:04 am under Events
No Comments