Table of Contents
What is Gromacs?
Why include Gromacs? What do you gain?
How can Gromacs be that much faster?
If it’s faster, will the participants get more points?
Why didn’t you use Gromacs in the first place?
What is OpenMM and how does it related to Gromacs?
As a participant of Folding@home, is there anything I need to do?
What modifications have you made to Gromacs?
Isn’t Gromacs a GPL’d code? Where’s the source for your mods?
Why don’t you just use the GPL version? Why the special license?
What do the Gromacs developers think about all of this?
“Gromacs” is an acronym for GROningen MAchine for Chemical Simulations, and it is a free and open-source molecular simulation engine. Gromacs is extremely fast, and has been one of the main simulation tools in Folding@home for almost a decade. It excels at simulating proteins and lipids, exactly what we need for FAH.
For the scientists using Folding@home, the results have been dramatic. For certain calculations, it is 20 to 30 times faster than Tinker, our previous code developed by Jay Ponder. This has substantially increased the power of Folding@home. Since we have serial aspects of our simulations, getting results back quickly is very important. The latest tools in Gromacs make it much easier to quantitatively estimate the interaction strength of small molecules with proteins. This should make it possible to predict how effective proposed new drugs will be.
Gromacs is built for speed. Everything about it has been optimized to be the very fastest MD code on the planet, and it is widely considered so. It contains algorithms creatively designed for speed. It’s innermost loops are manually tuned in Assembly, with SSE, SSE2, and 3DNow! multimedia optimizations available for supported architectures. Altivec assembly loops optimize it for both Linux and Mac OS-X. This can make it an order of magnitude faster than any other program – an amazing feat. For us to include all of these optimizations into our current scientific code did not seem a judicious use of our programming resources (why reinvent the wheel?) and we instead decided to collaborate with the Gromacs team. We have also extended GROMACS to support multiple core CPUs with SMP and to be able to process on high-powered GPUs, both of which increase the speed of computation very significantly over the original version that used a single CPU core.
Sorry, the points for each WU will still be determined by the CPU time of the benchmark machine, and since it will use Gromacs as well, your stats won’t be affected.
When we started coding the FAH project in the summer of 2000, using Gromacs was not a possibility. Apart for licensing issues, we decided to use implicit solvent models for several reasons (speed, low memory and low bandwidth requirements) and at the time implicit solvation models could not be simulated by Gromacs. Once we started to move to explicit solvation models, Gromacs became more applicable and we started to port our code over several years later. GROMACS has since been extended to run both implicit and explicit solvent models, obviating the need for alternate simulation software.
OpenMM is an open-source library for modern molecular simulation. It provides hardware independence through two API levels, allowing nearly any code to perform molecular modeling with minimal extra coding. It also emphasizes hardware acceleration, delivering performance comparable to code that has been painstakingly tuned to the hardware by hand. In 2010 we worked with the Gromacs team to introduce OpenMM support for Gromacs 4.5. Once Gromacs could run on GPUs, it saw up to an order of magnitude faster performance than CPUs for implicit solvent simulations.
No. For the participants of Folding@home, nothing new needs be done as the update is automatic. The FAH clients will download the Gromacs cores if the WU needs them. The binary should automatically detect if SSE, SSE2, 3DNow! or Altivec optimizations are supported. There are no issues on Windows or Mac OS-X. For Linux, you’ll need a 2.4.x series kernel or later.
We work closely with the developers of the Gromacs software: reporting our modifications to the Gromacs team, while using some of the latest Gromacs features in our own code. The Folding@home developers have been working with the Gromacs development team (primarily Erik Lindahl) to incorporate Gromacs into Folding@home. Please go to the Gromacs.org web site to learn more about Gromacs and download the source if you’re interested.
Folding@home has been granted a non-commercial, non-GPL license for Gromacs, so we are not required to release our source. The copyright owners of any GPL code (in this case the owners are the Gromacs development team) can distribute the same piece of software with difference licenses in parallel. See the GPL FAQ for more info on this. However, we will release our patches GPL’d (see below).
Folding@home has special security checksums meant to maintain error detecting and data integrity. To ensure the validity of the work calculated, we need to keep these routines closed source. However, this is not relevant to other Gromacs users and all other modifications will be released GPL to the Gromacs developers.
Here’s a quote from Erik Lindahl, one of the primary Gromacs developers: “We just love to see the code put to good use, and we’re very excited to be part of the largest computational folding experiment in the history of science. We also believe it’s important to support Folding@home since it’s an open project in contrast to many of the alternatives; it makes a tremendous difference to publish the scientific results instead of hiding them in proprietary databases (and it’s only fair since you contribute the CPU time).”
For More Information, Please See:
Last Updated on July 01, 2013, at 10:55 AM