Log in or Sign up

F@H Updates

Discussion in 'Folding@Home - Join Team Guru3D !' started by PantherX, Jun 1, 2010.

Tags:

Page 13 of 16

iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
new site link http://folding.stanford.edu/home/blog

FAH VM server down, sys admins notified
January 2, 2014 by Vijay Pande ·

A key FAH VM server is down, which means that stats updates are on hold (but points are still being kept on the WS’s, just not reported to the web site) and also the GPU AS is down. We will post an update when we know more. Unfortunately, Stanford is shutdown this week, so the sysadmins are on a reduced schedule, so the response time will be longer than normal.
Click to expand...

iancook221188, Jan 2, 2014

#241
PantherX Folding@Home Team Leader

Messages:

1,380

Likes Received:

0

GPU:

Nvidia GTX 1080 Ti

From the F@H Forum:

abdulwahidc said:

These two servers will be intermittently from Friday Jan 3 through Sunday as our university is upgrading infrastructure and doing maintenance.

fahnd03: 129.74.246.143
fahnd04: 129.74.246.144
Click to expand...

PantherX, Jan 3, 2014

#242
PantherX Folding@Home Team Leader

Messages:

1,380

Likes Received:

0

GPU:

Nvidia GTX 1080 Ti

From the F@H Blog:

We see a shortage in Core17 WUs, so we’re building more core17 WUs right now. The building process takes some time, probably until tomorrow morning. By then, there will be plenty of Core17 WUs.
Click to expand...

PantherX, Jan 4, 2014

#243
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

VM server went down again — back up now
January 6, 2014 by Vijay Pande ·

The VM server went down again last night. It’s back up now, but I’m not happy that this has happened so frequently. We’re looking into what’s going on here. Later this week, we’ll have a redundant GPU AS which will minimize the impact of this machine going down, but still we need to find out the root of what’s going on here. It’s possible that VM server is hitting end of life and having some hardware issues.
Click to expand...

iancook221188, Jan 6, 2014

#244
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

An Introduction to Markov State Models
January 8, 2014 by Vijay Pande ·

MSMBookFor researchers curious about how FAH efficiently uses thousands to hundreds of thousands of processors, we’ve recently put together a new book which describes these Markov State Model methods in detail.

The aim of this book is to explain the importance of Markov state models to molecular simulation, how they work, and how they can be applied to a range of problems. The Markov state model (MSM) approach aims to address two key challenges of molecular simulation:

1) How to reach long timescales using short simulations of detailed molecular models

2) How to systematically gain insight from the resulting sea of data

MSMs do this by providing a compact representation of the vast conformational space available to biomolecules by decomposing it into states—sets of rapidly interconverting conformations—and the rates of transitioning between states. This kinetic definition allows one to easily vary the temporal and spatial resolution of an MSM from high-resolution models capable of quantitative agreement with (or prediction of) experiment to low-resolution models that facilitate understanding. Additionally, MSMs facilitate the calculation of quantities that are difficult to obtain from more direct MD analyses, such as the ensemble of transition pathways.

This book introduces the mathematical foundations of Markov models, how they can be used to analyze simulations and drive efficient simulations, and some of the insights these models have yielded in a variety of applications of molecular simulation.

https://www.youtube.com/watch?feature=player_embedded&v=oh-xqR79QM0

For those who are curious for a less detailed introduction, this on-line seminar from FAH team members is also useful:
Click to expand...

iancook221188, Jan 8, 2014

#245
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Filed Under: Uncategorized ·
Webinar: Learn about OpenMM, the code driving FAH GPUs
January 8, 2014 by Vijay Pande ·

For those that are curious about OpenMM, the code that drives the GPU calculations in FAH, please join us on January 16th for a webinar presented by Professor Vijay Pande. I’ll talk about about OpenMM, the computational engine behind the Folding@Home distributed computing system, and hear how others have leveraged its high-performance on GPUs and its custom classes and extreme flexibility to accelerate research in areas such as free energy calculations, protein folding, and protein conformational change.

This webinar is planned for January 16th 2014 at 9.00 AM Pacific Time. Space is limited, so please register at http://bit.ly/OpenMM
Click to expand...

iancook221188, Jan 8, 2014

#246
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Working with Apple on GPU support
January 9, 2014 by Vijay Pande ·

There’s not a lot to report just yet, but we’ve made some progress in terms of Apple engaging us on the bugs we’ve filed some time ago that are keeping us from supporting our GPU code (OpenMM) on OSX. No promises here, but at least there’s some progress.
Click to expand...

iancook221188, Jan 9, 2014

#247
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

New backup GPU Assignment Server update
January 9, 2014 by Vijay Pande ·

We’re happy to report that we’ve got a backup GPU AS online. As with any of our servers, you can say “hi” and test your connection to it manually by sending your browser to it: http://assign-gpu2.stanford.edu (you’ll see it say “OK” if you can get through).

The use of backup GPU AS is new to the FAH infrastructure and we now see that to make it work, we’ll need to roll out a new client. That has now become the top code development priority.
Click to expand...

iancook221188, Jan 10, 2014

#248
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Getting ready for the new GPU AS
January 10, 2014 by Vijay Pande ·

We’ve done some serverstat tweaks for the new GPU AS, namely to include a row for that AS (so donors can see if it’s up) and a column for WS’s that were assigned by it. We’re also planning to have a new client in beta testing with support for the new GPU AS (amongst other improvements) soon, hopefully entering beta testing on January 10, 2014 (California time).
Click to expand...

iancook221188, Jan 10, 2014

#249
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Backend work
January 10, 2014 by Vijay Pande ·

We’ve been doing work on the FAH backend servers. We had to take the stats down for a bit longer than we liked. They’re back up now. Sorry for the delay in getting those points recorded.
Click to expand...

iancook221188, Jan 12, 2014

#250
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Progress with OpenMM on OSX
January 11, 2014 by Vijay Pande ·

In working with Apple, we’ve been making good progress with OpenMM (our GPU code, the heart of core17) on OSX. While we can’t promise anything just yet, this was encouraging. For those who are curious, you can watch our development in the open (OpenMM is of course, Open Source software) at our github repo commits link:https://github.com/SimTk/openmm/commits/master
Click to expand...

iancook221188, Jan 12, 2014

#251
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Fighting cancer on Folding@Home: EGFR
January 14, 2014 by john Chodera ·

[Guest post by Daniel L. Parton of the Chodera Lab, Memorial Sloan-Kettering Cancer Center.]

We’re about to roll out our first major F@h project on the new work server at Memorial Sloan-Kettering Cancer Center (MSKCC). Our chosen target is a protein known as the epidermal growth factor receptor (EGFR), which is a member of the kinase family of human proteins. Kinases are key components in cellular signaling pathways, which control and coordinate all cellular activities. When kinases are activated, they pass messages to other proteins by attaching small phosphate molecules to specific parts of their structure. For proper regulation of signaling pathways, the cell must also be able to temporarily inactivate kinases. However, mutations in kinase genes may cause them to become constantly active, resulting in overactivation of the associated signaling pathways. In some cases, this causes the cell to begin dividing and growing uncontrollably, and this chain of events is the cause of many types of cancer.
EGFR mutations are specifically associated with certain types of cancer of the lung, breast, colon, rectum, and brain. Although a small number of anti-EGFR drugs are available (for example, gefitinib, erlotinib, and cetuximab), the development of resistance mutations during treatment is a huge barrier to their effective use.

The Chodera Lab is working to understand a number of factors which have historically hindered the development of effective kinase drug treatments. Firstly, we are interested in what makes a drug molecule selective for one kinase over another (since there are approximately 500 types of kinase in a human cell with nearly identical active sites). Secondly, we want to understand the physical mechanism by which resistance mutations develop to diminish the therapeutic effect of a drug. Eventually, we also hope to be able to use genetic sequencing data from individual cancer patients to predict which particular drug treatment might be optimal given a particular patient’s tumor mutations. There are many theoretical and algorithmic hurdles that we will need to overcome on the way, but these F@h simulations of EGFR will be a first step towards these goals.

Thanks again to all the F@h donors for contributing their computer time to enable this research!

–

The team of researchers studying EGFR includes Daniel L. Parton, Patrick M. Grinaway, Kyle A. Beauchamp, and Sonya M. Hanson in the Chodera Lab at MSKCC. Special thanks to Richard Knospler in the MSKCC Open Systems group for all of his hardware and software help in getting our servers up and running!

The epidermal growth factor receptor (EGFR) kinase catalytic domain bound to a small-molecule experimental inhibitor. PDB accession code: 2HZ0. Image by Daniel L. Parton.
Click to expand...

iancook221188, Jan 16, 2014

#252
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Revised plans for BigAdv (BA) experiment
January 15, 2014 by Vijay Pande ·

We’ve put a lot of time into reading the comments about the BA experiment and have come to some conclusions regarding how we should proceed. BA was originally conceived as an experiment to push FAH as close to what you could run on a traditional supercomputer as possible, doing calculations that most researchers thought could never run on a distributed computing platform. In order to make this possible, the requirements for BA would have to be pretty extreme and constantly updating (much like how supercomputers are constantly being updated to the latest hardware). In recognition of this extreme set of requirements came a very large PPD.

BA was embraced by donors with powerful machines. It also encouraged donors to buy and build powerful machines, which are naturally expensive, in turn leading to donors naturally becoming upset when requirements change. However, updating BA is important. Without updating to keep BA amongst only the top machines in FAH, there would be a huge point inflation (turning off non-BA donors) and also limit what we could do with FAH outside of BA-needed projects.

In the most recent announcement of an update in BA, the reaction was particularly negative. This was far from our intent — we’re here to push our research forward and also to help bring donors together in this important cause we’re all fighting for. The BA update did the opposite. It turned people off from Folding@home and served as an obstacle to our ability to push our research forward.

We’ve listened to the donor comments (including those running BA and those who are not) and come to the following plan to be sensitive to their concerns but also to avoid this sort of issue continuing over and over in the future.

1) The posted change in BA requirements will be revised. The only change in requirements going forward will be to require 24 cores (with according changes in deadlines) and that will occur on May 1, 2014. Why change the minimum core count? In general, FAH works best when we have fewer longer trajectories rather than more slower trajectories, which is why we need to change the BA requirements periodically. Given the amount of BA work we would like to do, the cutoff has to be raised above the current level. While, 32 cores would be ideal but we can still get work done at 24 and, in recognition of donor needs, we will set the level there.

2) The BA experiment will permanently end on January 31, 2015. On that date, the servers will be set to accept only and we will have no plans for future BA WUs. This would allow donors to continue to use their machines and recoup more of their investment than the previous plan. This decision would also work to avoid future issues and strife within the FAH community associated with BA. We understand that many donors will be very disappointed about this decision. This was a judgement call I had to make and this decision is, in my opinion, the right thing to do for the long term good of FAH, even though I know there will be many upset donors right now.

In leading FAH, my approach has been to push the limits, try new experiments, but also keep an eye to the future such that FAH outlasts and out performs other distributed computing projects. Starting with establishing FAH itself over 13 years ago, to pushing to GPUs, true SMP, Playstation 3, and most recently to supercomputer like nodes with BA, we have constantly been trying new directions to see how we can further and further advance our research. All experiments come to an end, sooner or later. I think my team and I have learned a lot from reading the recent posts and it’s time for us to concentrate on the core parts of FAH and improve them and not bite off too much.

Going forward, the next steps will include a discussion of the change of the QRB formula and possibly an update of the benchmark machine. Our plan is to seek more input from donors for both changes. While a distributed computing project cannot be run effectively through polls, I think there is a lot of room for us to improve in terms of connecting to donors and incorporating their concerns. I’m very excited about the future of FAH. I think my team has learned a lot with BA and hopefully we can take that going forward to make FAH even better.
Click to expand...

iancook221188, Jan 16, 2014

#253
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

PG Member Paul Novick Thesis defense: drug repurposing successes so far
January 16, 2014 by Vijay Pande ·

Today is an exciting day, as Paul Novick (one of the Pande Group members working behind the scenes, from a FAH perspective) is giving his thesis defense today. The title is ”Computationally Guided Re-purposing as a Drug Discovery Strategy”. For those in the Stanford area, it will be in the Alway Building, M114 (Med School).

As a teaser and intro, here’s an intro to what he’s been working on:
Over the past several decades, the ability of the pharmaceutical industry to develop novel, impactful new therapies has diminished to the point where many formerly industry leading companies are now facing unsustainable business models. Development of a new chemical entity from initial target identification is now estimated to take 14 years with a price tag approaching $2 billion dollars. With many revenue driving therapies coming off patent within the next few years, there are serious concerns about the ability of the pharmaceutical industry to continue to finance innovation and the development of promising new drug candidates. In order to maintain solvency, heavy investments have been made in research technologies – such as genetic sequencing and analysis, high throughput screening, and combinatorial chemistry, among many others – promising to deliver faster, cheaper development timelines. Unfortunately, these approaches have yet to yield the dramatic improvements in research efficiencies that the industry so desperately needs.
Paul will talk about the solutions we have developed to tackle these issues, by combining computation and drug repurposing. This has yielded novel treatments for Chagas Disease and Dengue fever.
Click to expand...

iancook221188, Jan 17, 2014

#254
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Update on PPD calc in latest v7 client
January 17, 2014 by Vijay Pande ·

We’ve gotten donor feedback regarding issues with the Points per day (PPD) calculation in the latest client v7.4.0 beta. Here’s an update. After a code review, we see that v7.3.11 used the timeout (aka preferred deadline) not the deadline (aka final deadline) in the bonus calculation. This is now fixed in the source and will be rolled out in an update.
Click to expand...

iancook221188, Jan 19, 2014

#255
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

CPU counts, FLOPs, and corporate partnerships
January 20, 2014 by Vijay Pande ·

One question that has come up recently is that the count of the number of CPUs involved in FAH has gone down quite a bit in the last 3 to 6 months. There are several factors going on here.

We frequently negotiate periods of time to run on the machines in corporations, as this can bring a lot of power to FAH. In most cases, the donations we get from large corporations are for pre-set lengths of time. These companies have some campaign or some period of availability. We’re constantly negotiating with companies for such partnerships and so the fluctuations in the number of computers may be quite great, especially as the number of single-core clients changes rapidly.

Also, in recent partnerships, GPUs and large-core CPUs were emphasized over lots of single core CPUs and so the count of the number of machines will likely go down quite a bit but the FLOPs will remain quite high (today we’re still about 15 Petaflops, which makes FAH a very, very powerful supercomputer, especially when you consider that other traditional supercomputers would underperform their reported FLOP count if they used FAH’s approach at calculating FLOPs). In the end, a high FLOP count has a big impact on our work.

With that said, it is very important for me to emphasize that we value all contributions. Indeed, we have in the past (and will continue to show in the future) shown the power of how many computers, even single core PCs, can band together to do amazing things. Towards those ends, we have in particular been working on a new server backend software which can more efficiently utilize the results from many single-core CPU machines.

As I see it, the future of FAH will have to be heterogeneous and may even in the future incorporate mobile devices. In the coming weeks, we hope to roll out our plans there more publicly, with some new initiatives on the CPU side which I think could greatly increase the power of FAH.
Click to expand...

iancook221188, Jan 22, 2014

#256
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Watch FAH/PG code development in action
January 21, 2014 by Vijay Pande ·

Donors have often asked how they can see more of what’s going on behind the scenes. One way to do that is to check out the various code development projects we have going on at github. You can see some of the key tools being developed at https://github.com/SimTk, including OpenMM (the software behind Core17) and MSMBuilder (the software behind how FAH analyzes WUs).
Click to expand...

iancook221188, Jan 22, 2014

#257
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

NVIDIA GPU Technology Conference (GTC)
January 23, 2014 by Vijay Pande ·

I’ll be speaking at the NVIDIA GTC on our recent work in cancer. Specifically, it’s talk S4133, entitled “OpenMM Molecular Dynamics on Kinases: Key Cancer Drug Targets Revealed with New Methods and GPU Clusters” and it is scheduled for 50 minutes on Tuesday, 3/25/2014 at 2:30:00 PM in Room LL21E.
Click to expand...

iancook221188, Jan 25, 2014

#258
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Papers page updated
January 24, 2014 by Vijay Pande ·

We’ve added a few papers to the Papers page. The most important application result is our work on GPCR intermediates as new possible drug targets. We also have several new methodological advances which are proving important for our work going on right now.

Coming up (hopefully) soon will be a paper on new drug targets involving kinases, key protein targets involved in fighting cancer.
Click to expand...

iancook221188, Jan 25, 2014

#259
iancook221188 Ancient Guru

Messages:

1,729

Likes Received:

20

GPU:

GTX 670 SLI / GTX 460 SLI

From the Official FAH Blog:
http://folding.stanford.edu/home/blog

Kinase paper accepted in Nature Communiations
January 28, 2014 by Vijay Pande ·

Our recent work on simulating kinases on Folding@home and the simulation’s use in identifying potential new cancer drug targets has been just accepted for publication in Nature Communications this morning. I’ll post more in the coming weeks with more details after the paper is past Nature’s embargo period.
Click to expand...

iancook221188, Jan 29, 2014

#260

(You must log in or sign up to reply here.)

Page 13 of 16

Share This Page