Folding Forum

Hi guys

Is there a posibility to monitor the progress of a project? In my understanding, a project has a total number of workunits and after all workunits are done, the scientist have the results of this project and can use these results to do their research. Is there a way to see the progress of a project, like the total number of workunits, the number of finished workunits and the number of workunits left from this project?

Thanks

This doesn't exist, although I like the idea!

yeah, sounds really interesting! although it sounds like a lot of "busy work" that doesn't move science forward for the researches

i wonder, does anyone know how the PRCG numbering progresses? I assume right to left, like normal counting numbers, ie P123 Rx Cy Gz0...zn when all the Gen's are finished, it ticks over to P123 Rx Cy+1 Gz0

A project number is assigned to each research project. Runs and Clones denote variations in the starting conditions. Successive Gens are continuations of a single trajectory from the specified starting point.

If there are, say, 100 Gens of P,R,C then your WU represeents 1% of the overall trajectory and other people worked on the same trajectory, helping to lengthen it.

* Some projects need lots of starting points to span the entire N-dimensional space. (See the second figure for MSMs on https://en.wikipedia.org/wiki/Folding@home)

* Most projects benefit from longer trajectories.

Here's the definition of PRCG:
PRCG - Project Run Clone Gen, is used to identify a WU that is assigned to you.
Project: Biological system.
Run: Different starting configuration (positions) of atoms.
Clone: Different starting velocities of atoms (each Run can have multiple Clones with different starting velocities).
Gen: Time sequence (so Clone 1, Gen 2 is sent after Clone 1, Gen 1 is received and processed).

Based on that, we don't know how long it takes for the protein to fold to it's final state (or the several interim states that it may have). Thus, we can't say how many WUs are needed to begin with. The researchers regularly check the various Runs and Clones to see how many nanoseconds have been generated and see what data has been generated so far.

Some Projects can generate meaningful and sufficient data in months while others can take years... that's the nature of molecular simulation.

So, the researchers don't know, how much computing work has to be done in the first place, so they let the donors generate as much data as possible to get even more precise results over time. Thanks for the explanation

OK, I understand that the total number of workunits required to complete a project is unpredictable, making it impossible to track progress to the end target until you've got there. Still, it would be interesting to show the number of WUs completed for each project. It would help give a sense of progress of the collective community effort, not just one's own.

OldMax wrote:So I understand that the total number of workunits required to complete a project is unpredictable, making it impossible to track progress to the end target until you've got there. Still, it would be interesting to show the number of WUs completed for each project. It would help give a sense of progress of the collective community effort, not just one's own.

Or the number microseconds simulated so far ...

Now we have at least one data point for the duration of one project, and that was 100 ms of total simulated time, corresponding to several tens of millions of work units. I still expect others to be different by orders of magnitude, i. e. some might produce meaningful result from just 10 microseconds of simulated time, while others might conceivably require a whole second. Which is just about feasible.


Cheers,
HG

There's a 3rd party app being developed that may give you the information you're looking for. Unfortunately, running it is EXPENSIVE in terms of server resources so it's use either needs to be prohibited or it has to be deeply integrated with the stats server.