Upcoming changes to bigadv threshold

Any announcements about FAH policy, servers and new projects will be made here.
Moderators: Site Moderators, PandeGroup
 

1 post • Page 1 of 1

 
Upcoming changes to bigadv thresholdby kasson » Tue Dec 17, 2013 9:43 am
We have a policy of periodically re-evaluating the bigadv program, including the threshold required to run bigadv projects.
It is the intent of bigadv to match large and resource-intensive work units with some of the most powerful machines used by FAH donors. This "most powerful" line naturally advances with computing power. To date, bigadv has been a CPU-based program, and with increasing numbers of CPU cores and power of those cores, we have decided to lay out a roadmap of bigadv threshold changes for the next several months.

Feb 17 (two months from today): bigadv threshold will become 24 cores
Apr 17 (four months from today): bigadv threshold will become 32 cores

We want to give advance notice of these changes, and we recognize that change is not always welcome or comfortable. We should also emphasize that the science performed by donor machines is valuable in all parts of the FAH project, and part of the change in bigadv threshold is because we would like to encourage moderately powerful machines to help boost the capabilities of non-bigadv SMP projects where we do a lot of this science.

We also recognize that core count is not the most robust metric of machine capability, but given our current infrastructure it is the most straightforward surrogate to evaluate.

Thank you once again for your generous participation in the Folding@Home project!


kassonPande Group Member Posts: 1864Joined: Thu Nov 29, 2007 3:37 pmPrivate message This will kill the "ankle biters"

Well, at least I've been around Folding long enough to know that it is a matter of "when" (not if) changes like this will come.  This change will finally eliminate my SR-2 from being able to do Bigadv, but it serves many other functions anyway (including Crunching which it does very well).  The main thing is not to allow this news to adversely affect your DC interest because it is part and parcel of Folding.  Use the rigs that won't qualify for Bigadv for either SMP Folding, Crunching or both.  Fold and Crunch on my Bros and Sis'!! 

Sucks that they keep upping the core count needed for those of you already invested, hopefully they bump the points up even higher for those of you who meet the guidelines otherwise I find it hard to see the point.  This is why I am hesitant and don't really want to build a BigAdv rig myself just because it may soon become obsolete.
 
From the sidelines here are my questions for those of you with these BigAdv rigs.  Aren't the Intel CPU's more powerful than the AMD ones with fewer cores than the counterpart?  I mean that seems like a slap to me.  Doesn't the higher core count CPU's need to be overclocked most of the time just to finish the WU in time?  Seems that core count isn't necessarily all that's needed, but a decent CPU speed whether stock or not is needed to maintain a clean finish?  Finally, we aren't pushing core counts as fast as we were years ago, so how high are they going to keep upping the core count for BigAdv when clockspeed naturally drops as a result because I'm sure hyper-threading doesn't count, or does it?
 
Still trying to get an understanding of what you BigRiggers go through, it just seems so difficult for the investment.  I remember it wasn't too long ago they got rid of 16 core BigAdv.

How long does it take to finish a bigadv project on a 780ti? 
Bigadv is only on CPU at this time. There was some talk about GPU bigadv but no time frame. This is one way to cure running out of work in bigadv. At least my electric will drop back to normal (along with points).

zildjian75
What's this crunching thing I keep hearing about?

You ask and so you will receive read here:
 
http://forums.evga.com/tt.aspx?forumid=79
 
That's the Crunching@EVGA team thread.
 
https://secure.worldcommunitygrid.org/index.jsp
 
This is the WCG forum... currently popular work units right now that are running full time are Fighting Aids @ Home and Mapping Cancer Markers.
 
WCG and the BOINC client can do a lot with machines that no longer do the bigadv and you still get to contribute to what may one day be a solution to a health problem.
 
Stop on by the other side... many of us fold and crunch... I'm one of them. 

Thanks viper... I will check those links out.

bison88
Sucks that they keep upping the core count needed for those of you already invested, hopefully they bump the points up even higher for those of you who meet the guidelines otherwise I find it hard to see the point.

I'm hoping that the reason for this new tightening is due to new projects in the pipeline that are even more resource-intensive.
 

bison88This is why I am hesitant and don't really want to build a BigAdv rig myself just because it may soon become obsolete.

They don't change the minimum core count that often; the last time they did it was 2 years ago.  If you build something now that is above the bare minimum, chances are pretty high that it'll be supported for a couple more years.
 

bison88From the sidelines here are my questions for those of you with these BigAdv rigs.  Aren't the Intel CPU's more powerful than the AMD ones with fewer cores than the counterpart?  I mean that seems like a slap to me.

Linux and Windows don't make a distinction between threads and physical cores.  To both of them, an i7-4770k appears to have 8 cores.  Stanford's servers use the OS's reported core count when sending out work, so a 16 core/32 thread Intel machine will be able to get bigadv work after these changes.
 

bison88Doesn't the higher core count CPU's need to be overclocked most of the time just to finish the WU in time?

I didn't think so.  People just do that because they can, and the high point values of bigadv mean that overclocking has a much higher effect on PPD.
 

bison88Seems that core count isn't necessarily all that's needed, but a decent CPU speed whether stock or not is needed to maintain a clean finish?  Finally, we aren't pushing core counts as fast as we were years ago, so how high are they going to keep upping the core count for BigAdv when clockspeed naturally drops as a result because I'm sure hyper-threading doesn't count, or does it?

 
The clockspeeds haven't been dropping that much, and more cores makes up for any lowering of the clockspeed assuming you're not slashing it by, say 75%.

So glad my $800 investment will be non-useful less than 1 year after I purchased them :(
 
 
Actually, I am going to ask a friend to see if he can write a program to simulate both 24-core ankle-biters as one 48 core machine. If he can't, then I will be going Office Space on these things!

I'll have to pass on that. But if this announcement angers you, bring it up on Foldingforum.
 
Let your voice be heard.

bill1024
Maybe if everyone turned off all FAH for 48-72 hours call it being on strike.
They will get the hint.
My hexcore AMD-1045T was o/c to 3.6ghz and used to get 18k or so on regular smp.
Now I am lucky to get 10k ppd.
I would not mind doing smp, but why the big reduction if they need more to fold them?

I'm on board with this!!! Maybe we need to shut the evga team down for a few days....

Now that I think about it...  I have been folding for 5 yrs...  Thousands of dollars in hardware not to mention the electricity.  I have not seen one bit of any results from this.  I'm not saying they don't exist, but does anyone have a link to some "discovery" bring made from all this work or are we all secretly mining bitcoins for them. 
 
Just saying they get alot of free horsepower for running their operation, but keep shooting the hardware donors in the foot. It starting to remind me of alot of other not for profits...  Justa jobs program, while very little "work" gets accomplished.
 
Don't anyone take this the wrong way...  but they are valid questions!!  They seem a little ungrateful from someone getting millions of dollars of work done for free.

zildjian75
Now that I think about it...  I have been folding for 5 yrs...  Thousands of dollars in hardware not to mention the electricity.  I have not seen one bit of any results from this.  I'm not saying they don't exist, but does anyone have a link to some "discovery" bring made from all this work or are we all secretly mining bitcoins for them. 
 
Just saying they get alot of free horsepower for running their operation, but keep shooting the hardware donors in the foot. It starting to remind me of alot of other not for profits...  Justa jobs program, while very little "work" gets accomplished.
 
Don't anyone take this the wrong way...  but they are valid questions!!  They seem a little ungrateful from someone getting millions of dollars of work done for free.

http://folding.stanford.edu/home/papers

Whelp...  I just figured it out!!!  Only from the genius minds of academia.
 
Academia #1  "We need less folks folding BigAdv and more folding SMP."
 
Academia #2  "Alright.  Let's just raise the requirements for folding BigAdv."
 
Academia #1's 6th grade child - "Why not raise the payout of the SMP work unit to make the SMP more attractive while retaining the horsepower provided for BigAdv and therefore not alienating the ones who care and help our cause the most."
 
*silence*

Zagen30

zildjian75
Now that I think about it...  I have been folding for 5 yrs...  Thousands of dollars in hardware not to mention the electricity.  I have not seen one bit of any results from this.  I'm not saying they don't exist, but does anyone have a link to some "discovery" bring made from all this work or are we all secretly mining bitcoins for them. 
 
Just saying they get alot of free horsepower for running their operation, but keep shooting the hardware donors in the foot. It starting to remind me of alot of other not for profits...  Justa jobs program, while very little "work" gets accomplished.
 
Don't anyone take this the wrong way...  but they are valid questions!!  They seem a little ungrateful from someone getting millions of dollars of work done for free.

http://folding.stanford.edu/home/papers

 yup...  Now I understand why they can't handle public relations to the donors.  Trying to figure all that out leaves no time for common sense.

Just read thru that thread... interesting... A lot of what they see is what I feel.  However, given PG's history this will happen.  They (PG) really don't listen to donors.  We are a mere inconvience to them and should just fold and shut up so they can figure out the science.