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)

Computation errors and GPU temperature : some numbers

By , the in Hardware - No comment
This is something we knew, but without being able to define the real impact. Our GPUs are more likely to produce computation errors than our CPUs. However, temperature seems to be a decisive variable in the growth of error probability. This news (in French) shows a small panel of nVidia cards from various architecture and the testing conditions.

Of course, the cards in desktop cases are the main focus, but the laptop or high performance computing rack cards are more likely to be subject to these issues because they run in a much more confined environment. The good news is that most modern GPUs (Fermi or Kepler) are kept away from erroring by efficient thermal protections.

This experiment consolidates our usual advice : avoid folding on laptop GPUs, and make sure that your sensitive components are correctly cooled.

Source : Tom's Hardware France

Introduction of Vincent Voelz laboratory at Temple University

By , the in Folding@Home Project - No comment
Introduction of Vincent Voelz laboratory at Temple University
The Professor Vincent Voelz laboratory has started in August 2011 at Temple University in Philadelphia, PA. Two servers for the Folding@Home project have been set up there and the first simulations hosted on them have started this summer. In the meantime, the team had an acces to high performance computing cluster of Temple university Institute for Computational Molecular Science to generate some initial data for these simulations.

One of the main goal of this laboratory is to use molecular simulations for computational design of folding and binding properties. This design requires folding for lots of different possible protein sequences, which is a natural task for the Folding@Home distributed computing platform. Vincent Voelz’ team works to consolidate Markov State Models of conformational dynamics to do efficient estimation of the effects of sequence perturbations. A good starting point to test these effects are to look at proteins for which many sequences have been characterized, to see if it is possible to predict sequence-dependent changes. Many of these sequence mutations are important in human diseases, so professor Voelz hopes to gain insight into these process as well.

High performance computing cluster at Temple University.

Source : Vijay’s blog

Understanding the folding of hIAPP, the peptide linked to the Type 2 diabetes

By , the in Folding@Home Project - No comment
Understanding the folding of hIAPP, the peptide linked to the Type 2 diabetes
Here’s an update from the Professor Xuhui Huang’s laboratory at University of Sciences and Technology of Hong Kong, another laboratory collaborating inside the Folding@Home consortium.

In addition to the study of molecular recognition process, another goal of this laboratory is tu use the Folding@Home platform to explore the folding of free energy landscape of the human islet amyloid polypeptide (hIAPP). hIAPP, also called amylin, is a 37-residue peptide and its aggregation reduces working beta-cells in patients with Type 2 diabetes. As an intrinsically disordered protéin, the hIAPP monomer doesn’t have a folded global minimum in its folding free energy landscape, but contains many stable local minimums. Thus, understanding these local stable states can help us to understand the amylin aggregation mechanisms and then design some small molecules to inhibit the amyloid formation.

As we have seen in Vijay Pande’s laboratory simulations on the alpha beta peptide involved in Alzheimer disease, this research may lead to potential therapeutic agents for the Type 2 diabete. On the Folding@Home platform, professor Huang’s team is running extensive simulations on molecular dynamics (MD) and they are building Markov state models to elucidate the free energy landscape of the hIAPP monomer. Projects 2974 and 2975 are related to this study. The laboratory team wishes to thank Folding@Home donors who make these research possible.

Source : Vijay’s blog

Receptor binding by the Influenza virus

By , the in Folding@Home Project - No comment
Professor Kasson's team has recently published an article in the Biochemistry journal detailing how the influenza virus binds to cell surface receptors. In this article, the researchers detail how simulation techniques can be used to further analyze the biochemical and structural data of glycan binding by the influenza virus. They also review the work done in collaboration with Professor Pande's team, including research completed using Folding@home. Previous articles on the same subject can be found here and here.

20 glycan molecule structures are connected to the receptor-binding pocket of hemagglutinin.

Source: Vijay's blog

FAH simulations lead to a new treatment strategy for Alzheimer's Disease

By , the in Folding@Home Project - 1 Comment
We are happy to finally announce some very important results from the project. These results have been a long time in development and are great achievements for Folding@home in general, showing that the approach started 10 years ago can significantly advance the long-term goals of the project.

Those long-term goals are:

  • 1) Develop new methods to overcome the challenges presented by computer simulation of protein folding
  • 2) Apply these new methods to better understand the process of protein folding
  • 3) Use these new methods to simulate the misfolding of the protein "beta amyloid," a key process of toxicity involved in Alzheimer's disease
  • 4) Use these simulations to develop new small molecule drug candidates to treat Alzheimer's Disease

In the early years of FAH, the focus was on goals 1 and 2. In the last 5-7 years, the focus changed to goal 3. Now it is time to announce some progress on the final goal!

In an article published in the Journal of Medical Chemistry, researchers describe testing predictions made by previous work done by Folding@home, and how these predictions have led to a new strategy to combat Alzheimer's Disease. While it is true that this is not the cure to the disease which we all seek, it is a major step towards that cure.

The next step, currently underway in the laboratory, is to test this new compound in the hope that it will be able to be turned into a usable drug. It is too early to discuss preliminary results (as those have not yet been peer-reviewed), but Vijay is pleased that the guidelines set out in the article do point to a viable drug, not just a candidate!

Article: Design of β-Amyloid Aggregation Inhibitors from a Predicted Structural Motif.

Source: Vijay's blog

Switch to core a4 for more points!

By , the in Folding@Home Project - No comment
Stanford definitely wants us all to move over to using Core a4, and they know full well how to motivate us to do so.

Core a4, successor to cores a2 and a3, is the first that works equally well in both single- and multi-core modes. However this feature only works properly for v6 clients 6.34 and higher, and all versions of the v7 client. Currently the a4 core is used for the most modern projects, with core a3 being kept for projects released before core a4 was considered stable.

In order to incentivise upgrading to use the new core, Stanford is using the ultimate carrot: a 10% points bonus on all units using the a4 core. This will not be visible on monitoring software as the project summary will not be updated with the boosted points, but will be applied by the servers nonetheless.

So if you have not already, switch to v7 for more points!

Source: Vijay's blog

Xeon Phi: an x86 coprocessor, but what about OpenCL?

By , the in Hardware - No comment
A few weeks ago, Intel restarted the hype about its abortive "Larrabee" architecture. This time the project is proceeding to the market, as proved by Intel assigning it a trade name: "Xeon Phi".

If you followed the announcement you will be aware that the new chip contains "more than 50 cores" capable of running x86 code, and up to 8GB of memory, all mounted on a PCI-express card rather than the more traditional CPU socket, to negate the need for a new design of motherboard.

Clearly this has changed the aim of the project from a GPU that is capable of GPGPU tasks, to that of a coprocessor, similar to the x87 chips of old.

The card is designed to work alone, or in a cluster as part of a wider supercomputer arrangement.

In terms of software support, details are currently unclear. The card will run a Linux OS, isolating itself completely from the rest of the system (one could ask how it functions as a coprocessor given this restriction...) and Intel's own marketing documents do not mention OpenCL. Is Intel ignoring this standard-in-the-making?

A few days ago, the Parallelis website scrutinized the documentation for the card and raised some questions about its effectiveness, particularly when running x86 code, and the limitations imposed by using PCI-express (which is much slower than interconnects between CPUs on the same chip, or CPUs and RAM, for example).

In short: this is a potentially fascinating development, but there are unanswered questions. Intel, the ball's in your court...

Source: PCInpact (in French)

Reducing side effects of IL-2 protein used to treat cancers

By , the in Folding@Home Project - No comment
Today we are highlighting the work of Chris Garcia's laboratory at Stanford Medical School. The laboratory team had a great idea to re-engineer a very important protein used to treat cancers and Vijay Pande's laboratory got involved in the research by providing computer simulations that help to understand the mechanism by which the new protein worked. The results are exiting. You can access to the full article at this address:

The utility of a naturally occurring protein given, sometimes to great effect, as a drug to treat advanced cancers is limited by the severe side effects it sometimes causes. But a Stanford University School of Medicine scientist has generated a mutant version of the protein whose modified shape renders it substantially more potent than the natural protein while reducing its toxicity.

Source: Vijay's blog

Ready, set, Fold! Final Folding@Home v7 client and new website released!

By , the in Folding@Home Project - No comment
Ready, set, Fold! Final Folding@Home v7 client and new website released!
About one year after the public beta release, the v7 client (7.1.52) is finally considered as stable enough to become the official Folding@Home client. The main new features are:

  • Client code rewritten from scratch
  • Multi platform client (unified GPU, SMP and uniprocessor)
  • Centralized interface to manage all running cores.
  • A new viewer that should work better.

This release has been synchronized with a full redesign of the project official website. The old academic looking website is replaced by a much more modern one which should be much more easy to use for public audience. The new site is now guided by three steps:

  • Ready: Discover Folding@Home, proteins, why to fold them and the most important, how to help.
  • Set: Client download (the recommended version is adapted according to the OS of the machine visiting the website)
  • Fold: An installation guide, some information about the monitoring (FAHControl) and the viewer. Access to different level of help (FAQ, technical documentation and Official Forum) and a whole section about points, statistics, ladders and third-party statistic websites.

The site should also display in your language automatically when you visit

The new website is simpler and more attractive so as the new client that would probably be of a great help to convert your friends and family to join the Folding@Home project. You don't have to be a power-user anymore, FAH is ready to be spread widely!

Source: Vijay Pande’s Blog