There was a 5.6 magnitude quake pretty close to the Stanford campus last night at around 8pm pacific time. All is fine with FAH it seems, but it did shake us up quite a bit. It's nice to have rackmounted servers securely mounted to those racks!
Dr. Edgar Luttmann has been a FAH group member for several years. His primary areas of interest are Alzheimer's Disease and new methods for describing water. In studying Alzheimer's disease, Edgar has contributed to FAH in several ways. He's been performing simulations of protein aggregation involved in Alzheimer's disease as well as helping understand the experimental work coming from the group (primarily NMR).
In his second research area, Edgar has been working with Jason Wagoner in the Pande group to develop better models of water. These models were originally conceived in order to have a model which was more computationally efficient, but turns out that they may be both faster (by about 10x perhaps in some cases, maybe more) but also it looks like they may be in fact more accurate as well. Finally, these new models are also very well suited to GPU's and PS3's. Bottom line, we're very excited about this and hopefully will be submitting our first results for peer review soon.
For those that are local to the Stanford area and are interested in hearing more about Folding@home, I'll be speaking at the "Life in Motion" symposium this Thursday. You can learn more about it here http://biox.stanford.edu/news/symposiums.html . Here's a summary:
October 25th, 2007: "Life In Motion" Bio-X
has teamed up with Stanford's National NIH Center for Physics- based
Simulation of Biological Structures to hold a symposium entitled, "Life in Motion".
The goal of this symposium is to educate students and scientists from
different disciplines about the exciting uses of simulations driven by
the laws of physics and mechanics across a range of scales, from
molecules to organisms. The talks will be presented by a series of
experts and innovators from around the world.
The talks will be pretty broad in range (not just molecular and not just simulation), talking about motion in biology. BTW, "BioX" is a broad, interdisciplinary initiative at Stanford (and the Pande group is part of it).
While Adam Beberg is currently a grad student at Stanford in the Computer Science department and a member of the FAH team and Pande group, Adam has been collaborating with FAH from the very beginning. Adam has experience with distributed computing beyond FAH, as he was one of he founders of distributed.net and from his experiences with that, wrote the Cosm library to help with many areas, including writing distributed computing code.
In his thesis work, Adam has worked in many areas, including distributed storage (more on that later), distributed computing code, as well as the GPU code. For now, I'll highlight on his work with GPU's (and will comment on the rest at a later date). Adam has helped make major steps forward in our FAH code for GPUs, taking the code from "academic quality" to something which is very robust and could be broadly useful. This was a major undertaking, requiring knowledge in many areas including both the science, internals of GPUs, and the whole tool chain. With Mark Friedrichs (a programmer in the Simbios center and Pande group), our GPU code is looking great and we hope to be making some major announcements soon (although it's too early to talk about more details now).
We've been working to bring the rest of our web page into line with the new look. We've just updated the stats pages to have this new look. We still have a little work to do, but I think we're almost there.
The new web page is in place and all is looking pretty good so far and able to handle a pretty heavy web load so far. We still haven't had to deal with a slashdot scenario so far on the new site (perhaps the ultimate test), but I expect we'd do ok.
This is a radical redesign, so many of the links have changed. We've put some substitute links for the short term, but please make sure to switch over to the new links ASAP. With the new site, we should be able to keep the site up to date much more easily (and thus the site should stay more up to date), which should be a great help for FAH.
Dr. Chodera joined the lab about a year ago after completing his PhD at UCSF in Ken Dill's lab. John plays a role in many projects in Folding@home, but I'll highlight two primary projects today.
First, with former FAH team member Nina Singhal (now a professor at University of Chicago), John has further developed a new way to use distributed computing to simulate long timescale dynamics. This is at the very heart of how FAH works, so this is a very exciting advance for us. The basic idea is to build a kinetic model of the long timescale process by diving the system into a series of states and calculating the transition probability between these states via molecular dynamics. To make all of this possible requires some nifty statistics (especially a fair amount of Bayesian statistics, which is becoming more and more important to how we do our work). The upshot is that we should now be able use FAH to do calculations which we couldn't do before, which is very exciting for us and hopefully will have a big impact on the broader scientific community as well.
Second, John has been working to further develop methods to calculate free energies, especially those relevant for biomolecules such as protein-ligand free energies. These types of calculations are particularly relevant for our plans for late stages drug design, but also represent a major advance in computational methodology in general.
Check out this plot (also on our stats page). We see that over time, FAH has added about 40,000 active processors each year since FAH was released (in October 2000).
This graph doesn't talk about FLOPS, which have gone up dramatically faster, due to the advanced clients in FAH (SMP, PS3, and GPU clients). The FLOPS has gone up much faster, and we expect it to grow even more.
Note that these are *active* CPU's, not total "devices" or anything like that. Many projects like to list the total number of machines that have ever run the software (even briefly), whereas the numbers here are very different -- they are the number of machines which have recently returned results (and are thus "active").
We look forward to continuing growth in FAH, as we try to make FAH better and better than before.