News

Help, my CPU slot no longer folds !

By , the in Folding@Home Project - No comment
Help, my CPU slot no longer folds !
Recently, your CPU slot might have stopped folding with the following messages in the log :

Citation:
Connecting to 171.67.108.45:8080
Failed to get assignment from '171.67.108.45:8080': Empty work server assignment
Connecting to 171.64.65.35:80
Failed to get assignment from '171.64.65.35:80': Empty work server assignment
Exception: Could not get an assignment

In its default behaviour, the client always configure automatically the CPU slot to use fewer threads of the CPU than the total available :

  • if you have a GPU slot, the client reserves 1 thread to feed it
  • if you use little aggressive settings (Folding Power set to Light or Medium), the client will not use all threads

This behaviour usually creates a non even number of threads to be used for the CPU slot (if you have an 8 threads CPU, the client will attempt to use at most 7) which implies compatibility issues with a lot of simulated systems when this odd number is a prime number. This limitation comes from algorithms used by Gromacs (the science code) to decompose the system in parts computed by each thread. To determine the supported combinations, the researchers relied on returns from testers for each project and set up the assignment servers accordingly. Gromacs used to work well from 1 to 8 threads, then for every value that didn't contain a prime greater than 7 when factored (this means that all prime numbers greater than 7 were excluded).

Lately, the Folding@Home project started to follow official Gromacs recommendations more strictly, which exclude odd prime numbers greater than or equal to 7 from supported decompositions.

If you are in this situation with you CPU slot, the workaround is quite simple. You just have to change the automatic behaviour of the client (-1 value) to a forced setup (6 threads instead of 7 if you use an 8 threads CPU for instance) with the following method :
In FAHControl, click on Configure.
In Slots tab, select the CPU slot and click on Edit.
Near the top of the window, change the number of CPUs.
Click on OK then on Save.

If you need further assistance to configure your client, do not hesitate to post on our forum.

FAH-Addict is back !

By , the in FAH-Addict - No comment
FAH-Addict is back !
After more than two years offline, FAH-Addict.net finally makes a comeback !

Since two years, many things have changed on the Folding@Home project as well as in the Hardware world, so we expect a quick return of what has been the site's DNA :
  • Project news
  • Hardware news
  • Hardware reviews
  • Tutorials

We took the opportunity to modernize the CMS and its interface, so there might still be small slight issues with older contents. Don't hesitate to report the little bugs if you find some.

Ongoing maintenance on Folding@Home servers

By , the in Folding@Home Project - No comment
As mentioned on the official Twitter account :

A few key servers for Folding@Home project (the AS ones) are in maintenance for a few hours. Don't freak out, everything will be back online soon !

Source : Vijay's blog

EDIT : 27/08/2013 12:30pm UTC

Maintenance has been longer than expected due to network hiccups encountered after the servers were restarted. But today at about 2 am UTC, almost everything has been reactivated and things have been progressively doing better. Vijay Pande has officially announced it on Twitter after two intermediate messages :


You should keep in mind that some type of WUs are still causing issues : for instance, I’m still unable to send or get work for my old nVidia GPUs (pre-Fermi).

Source: Vijay’s blog

EDIT : 27/08/2013 8:00pm UTC

As we already noticed in previous update, VSP07 server, which handles WUs mainly for core 11, is still broken. System administrator of University are still working hard to fix it and put it back on line.


Source : Vijay's blog

FAHcon 2012 : Professor Peter Kasson

By , the in Folding@Home Project - No comment
During Folding@Home conference 2012, Professor Peter Kasson of University of Virginia presented his work on the influenza virus and he explained why this virus attracted so much attention.

First, the influenza virus kills about 40 000 people in the USA every year and many more worldwide. These are mostly children under 2 and adults over 60. This is obviously something that concerns us because everyone hopes to have children in good health and to live beyond the age of 60 … Second, influenza has proven track records of causing global mass-mortality, such as in 1918. A similar virus today could easily kill much more than 60 million people and we’d like to be prepared. Finally, the influenza virus is an interesting model for understanding other viruses such as HIV and cancer-causing viruses such as HPV (Human papillomavirus), Heptatitis C, and Epstein-Barr virus. It might surprise you, but many cancers are associated to a virus and these form an important area of prevention.

Folding@home has done a lot of work on how influenza virus enters into cells to replicate. This is an important therapeutic target and it is critical to understand why some viruses like H5N1 (“bird flu”;) have not become efficiently transmissible between people. Some of the recent work on the topic focuses on the folding of the cell membrane which is necessary for viral entry. Professor Kasson’s team obtained interesting results and we will post a news article when they are published.

The researcher also presented a promising new software suite: Copernicus. Peter Kasson’s, Eric Lindhal’s and Vijay Pande’s teams published a paper back in 2011 on the topic and they have pursued developments. Copernicus essentially makes the back-end control of large-scale simulations much more transparent, so FAH researchers will be more easily able to integrate new methods. This software also works on supercomputers and cloud computing platforms which makes it easier to run simulations that complement the simulations run on Folding@Home. It also allows other researchers from outside FAH to gain access to the kind of simulations we run on FAH. All these changes are on server side, so the donors should not notice any difference in the client behaviour. But thanks to Copernicus, we should see new research areas appear.

Source: Vijay’s blog

FAHcon 2012 : Doctor Greg Bowman

By , the in Folding@Home Project - No comment
Doctor Greg Bowman of Berkeley University showed a project at Folding@Home conference (FAHcon) focused on new therapeutics applications of proteins. It is known that the interleukin 2 (IL-2) protein can help to stimulate an immune response, so in theory, this protein might be efficient to treat patients with immune deficiencies.

However in practice, prescribing IL-2 to these patients often leads to severe heart problems. To find a better solution, researchers at Stanford University designed a variant of IL-2 that can stimulate this immune response without any side effects. However, they were unable to understand how this mechanism worked because the two proteins had almost identical structures!

Using Folding@Home, Doctor Bowman’s team showed that IL-2 is a relatively floppy protein while the one designed at Stanford is locked in a structure that is poised to stimulate an immune response.

Source: Vijay’s blog

Folding@Home Consortium Conference 2012

By , the in Folding@Home Project - No comment
On May 25th 2012 at Stanford University, the first Folding@Home Consortium scientific conference was held. The goals were to discuss about recent scientific advances, to share new techniques for how to better use the resources available to FAH as well as to plan Folding@Home platform advancements for the next year.

A series of news will follow to summarize the main ideas that came out of this conference.

Here is a picture of most participants:


Click on the image to enlarge.


From top left to bottom right: TJ Lane (Stanford), Dr. Jason Wagoner (Stanford), Prof. Dr. Vincent Voelz (Temple), Dr. Sidney Elmer (Sandia National Lab), Dr. Fancesco Pontaggia (Brandeis), Dr. Lan Hua (UCSF), Bruce Borden (FoldingForum.org), Joseph Coffland (Cauldron Development), Dr. Diwakar Shukla (Stanford), Dr. Lee-Ping Wang (Stanford), Steven Kearnes (Stanford), Kyle Beauchamp (Stanford), Dr. Greg Bowman (UC Berkeley), Dr. Relly Brandman (UCSF), Robert McGibbon (Stanford), Prof. Dr. Yu-Shan Lin (Tuffs), Prof. Dr. Matt Jacobson (UCSF), Prof. Dr. Jesus Izaguirre (Notre Dame), Prof. Dr. Vijay Pande (Stanford), Prof. Dr. Michael Shirts (University of Virginia), Dr. John Chodera (UC Berkeley/QB3), Prof. Dr. Peter Kasson (University of Virginia), Prof. Dr. Xuhu Huang (Hong Kong).
Not pictured: Prof. Dr. Chris Snow.

Source: Vijay’s blog

Peptoids

By , the in Folding@Home Project - No comment
Peptoids
One of the projects that Prof. Vincent Voelz’s laboratory is excited about is molecular simulation of synthetic polymers called peptoids. These are biomimetic molucules that can fold like proteins, but with different strucutral properties. Several peptoids that can fold in a unique three dimensional structure have been identified, but a better computational model is needed to identify the driving forces for folding and to be able to predict stable structures of these peptoids. If the laboratory is able to develop tools to achieve it, peptoids have the potential to become an amazing platform that can be used in all kind of applications, from biotherapeutics to nanomaterials.

So far, the researchers have shown that modern forcefields can accurately fold peptoids (results available on http://dx.doi.org/10.1002/bip.21575) and they are working with experimental collaborators on blind predictions of peptoid structures (new results will be published soon). Folding@Home would contribute to large-scale simulations of peptoid folding to better understand peptoid folding mechanisms and design principles.


Source: Vijay’s blog

Folding@Home 7.3.6 final released, a new website and a video !

By , the in Folding@Home Project - No comment
There is another revolution at Stanford : after a resoundingly successful release of v7 client and the complete website redesign, the 7.3.6 client is starting a new approach, aimed at simplifying the client for potential new donors. This release comes together with a new site design and a promotional video.

The client


The following main features have been announced by client developer, Joseph Coffland:

  • Web control interface added
  • Control buttons removed from FAHControl interface
  • Simplified installation

For the time being we haven’t been able to test these new features, but we’ll take a closer look.

The website


This is again a big change : the three frames which were part of the already simple interface are gone, replaced by a new home page completely dedicated to education with an animation oriented graphic style.


The website also emphasizes social networks (well yes, we’re not alone :hehe )

The text emphasizes the principles of protein folding, the opportunity to participate and introduces the network machines (us ;)) already in place.

The video


To complete this release, a Youtube video has been posted to introduce its audience to the implication of misfolded proteins in a huge variety of diseases, and it explains how to participate in finding cures for them !



Let’s communicate !


Stanford has offered us great tools to promote the project more easily, the time has come to recruit. Your family, your friends, the forums and the online communities you’re involved in might be interested ! It would be a shame not to enjoy this beautiful dynamic !

Main source : Vijay’s blog

FAH-Addict is also on social networks !

By , the in FAH-Addict - No comment
FAH-Addict is also on social networks !
We’re posting this little news to introduce you (or to remind for some of you) of our presence on the main social networks. Indeed, FAH-Addict is also a way to bring people to the project, because it doesn’t get huge coverage by the general media. Your help might be valuable in different ways:

On Facebook of course we have the biggest community. This is a good way to promote the project among your family and friends by sharing our news when you think they might be interested.

On Twitter which is our oldest community, you’ll find some other accounts interesting to follow according to your needs. We’d like to say hello to donors from other distributed computing projects that are following us ;)

On Google+ our latest and quickly growing community ! When sharing or adding some +1, you can help us reach the “featured” posts in regard to the user interests. This might be a good way to reach people in a targeted way.

Anyway, whether you’re a big social networks fan or not, thank you for your loyalty. Even when we’re not very active, we appreciate seeing you around on FAH-Addict !

GeForce GTX Titan – The most powerful single GPU card

By , the in Hardware - No comment
GeForce GTX Titan – The most powerful single GPU card
If you’re not shocked by spending a month of minimum wages on a computer component, you should know that nVidia is preparing to release a mythological monster !
The GeForce GTX Titan will be based on a GK110 chip, a development of the Kepler family, which is the first card on sale to use this component. The chip is supposed to be as powerful as 85% of a GTX 690 which is equipped with two high end Kepler chips !

The GK 110 chip is identical to the one embeded on the Tesla K20X GPGPU card. As a reminder, it is fitted with 15 SMX (2880 SP) with 14 activated (for 2688 SP). This chip is capable of delivering 3.95 GFLOPs on single precision calculations and 1.31 GFLOPs on double precision ones.

Here are some numbers, mostly as rumors, for the GeForce Titan GK100:

  • Released on February 18th
  • 2880 SP (15 SMX) with 2688 activated (14 SMX)
  • 7.1 billion of transistors
  • 502 mm² die
  • GPU frequency : 732 MHz
  • 6 GB GDDR5 on a 384 bits bus clocked at 1300 MHz
  • 235W TDP
  • Priced between $850 and $1000

In other words, a WU devourer ! Even if this kind of high end chip might be subject to early defects. nVidia is going to enforce strict design rules for this card on its partners, so no custom coolers!

The rumors about the availability date vary. The vast majority of the sources tends to point toward February 18th, though some of them mention the 25th …

Source : PC World (in French)