2 replies to this topic

[maestro949]

Many good topics at this year's SC06 conference. Below are a selection that will have some impact on the aging fight. There are PDFs for each presentation as well. Hopefully the videos for these find their way onto Google videos or Youtube.

Ray Kurzweil will be presenting on The Coming Merger of Biological and Non Biological Intelligence

They also have a new Exotic Technology Initiative where panels will try to predict what the HPC architectures of 2020 will be.

Jerry Bernholc will be presenting on Atomic Scle Design of Nanostructures. An interesting quote from the abstract... "We will show that key parts of our algorithms scale linearly with the number of atoms, making them ideal candidates for a petascale implementation." That's quite profound, especially if the same or similar algorithms can be carried over into biological sims. Efficient parallelization is one of the biggest challenges for HPC algorithms.

A presentation on the The Blue Brain Project

Theres a talk on Computational Approaches to Complex Ecological Networks which alone doesn't have any impact on aging research but future systems biology will likely rely heavily on models of the complex genetic network as the genomic and proteomic data comes online in these fields.

For the evolutionists and aging theorists there's PBPI: A High Performance Implementation of Bayesian Phylogenetic Inference

Locality and Parallelism Optimization for Dynamic Programming Algorithms in Bioinformatics

High-Performance Computing Methods for Computational Genomics

And the best for last. IMO the most significant long-term gains will be a result from significant advancements in Molecular Dynamics as it will have a significant impact on protein/drug design. Here, they talk about a new parallelized MD algorithm called Desmond that "achieves unprecedented simulation throughput and parallel scalability on commodity clusters." One impressive example of Desmond is "on a standard benchmark, Desmond’s performance on a conventional Opteron cluster with 2K processors slightly exceeded the reported performance of IBM’s Blue Gene/L machine with 32K processors running its Blue Matter MD code." Read the PDF for more.

More on Blue Matter Blue Matter: Approaching the Limits of Concurrency for Classical Molecular Dynamics

[arrogantatheist]

Thanks for posting these, I am going to enjoy reading through all those links.

[maestro949]

SC06 is underway and bloggers are, well, blogging...

scalability.org mentions Ray Kurzweil and MD-Grape. There's a reference to the Kurzweil video being uploaded but it's not up at the time of this post. The bit about MD-Grape is intriguing:

I walked by the MD-Grape bit. In a little SGI case. These units allow researchers to ask “what-if” questions, by supplying so much computational power to molecular dynamics simulations, that the research may be able to directly interact with the simulation, to change it, to measure things, to re-run with varying conditions. Now apply this to misfold diseases, or similar things of medical interest. It may be possible for researchers to find structural agonists/antagonists for misfold based disease. Multiple metabolic chemical pathways may be simulated under a variety of conditions, and again, it may be possible to use the computing power to design solutions or treatments for metabolic pathway diseases.

Mmm, that's good food for thought. Is there any resveratrol in them grapes?

ClusterMonkey blogs about Panasas' ActiveScale (think operating system for high performance clusters), programming tools and NVIDIA's C compiler for GPUs.

Windows Server Division WebLog talks about talking about grids and microsoft's entry into the clustering market


Check out the hi-def graphics generated in IBMs Interactive Ray Tracing Video and keep in mind, this is a computer generated image! Pretty amazing. PDF