Change in BA requirements

[kerryd]

We wanted a road map we got one for good or bad time will tell.I turned my 4p's back on at least tell May , then they go off for good I think only 16 cores each not much good for folding smp's.For the cost of cpus for them I can buy 3 gpu's that get 100k ppd each , and cost less to run.In May I will make up my mind what I will be doing with my computers tell then happy folding


Kerry

[DocJonz]

I'd like to thank Vijay and the team for listening to donor concerns - the candour (yep, English spelling!) is much appreciated. As with others, I was interested in a road map, and you have provided this - sure, some will be upset, but I think it is a sensible decision, and should prevent similar issues 'flaring' in the future.

(I am guessing that the comment "Going forward, the next steps will include a discussion of the change of the QRB formula ..." in the Blog might be there, in part, to ensure that high end machines, i.e. those that do BA now, receive a better return on SMP's than present? - that, at least, would make sense, but is pure speculation on my part ).

[orion]

Ill keep my 4p 24 core running until the deadline next year! and maybe beyond!

If your intent is to keep running BA that may be a problem. With the 5/01/2014 core count change there will also be a deadline change for the WU per Dr. Pande's post.

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.

So your 8431's may not make the BA deadline but if you're only running SMP then you should be good to go.

Expanding on this it would also be nice if we knew what the deadlines will be changed to.

That would give donors with 24 thread rigs and those with slow 32 thread rigs a better idea if they will be able to continue with BA after May 31.

Also if the deadline changes are extreme enough it could even bar some 48 thread rigs from doing BA.

[EXT64]

Might not know yet (might depend on how many parallel WUs (Projects/Runs/Clones) are left?). I'm more curious how an exact end date is given. Will the projects be done by then? Will some end before? Are we just in the realm of extra statistics at this point? (still valuable work, but not as exciting and something that you could shutoff whenever you wanted to).

[tear]

Might not know yet (might depend on how many parallel WUs (Projects/Runs/Clones) are left?).

Nope. Preferred deadlines have absolutely nothing to do with number of generations "to go".

Seeing the deadlines haven't been provided, Folding@Home is headed towards another bigadv issue (which
will arise the moment some 24 core systems are cut off due to the change of deadlines).

[texinga]

Hafta say that I was hoping for a different outcome for Bigadv than to see the project end in just over a year. I'd hoped that we could have some good working dialogue with PG and work-through the challenges that were shared in this thread. If we wore Dr. Pande down or otherwise brought them to a point where they felt they just could not manage this, then that is unfortunate and was not the desire of most Bigadv Folders that I know. It feels like that is what happened even though Dr. Pande has a positive and different future in mind. I still think that the collaboration between Scientists and Donors is a healthy thing to foster here. I hope that will grow and get better.

One last suggestion to Dr. Pande and his team. Don't ever be afraid or feel like you can't just talk to us while you are working on the future of Folding. Engage your donors while you are developing ideas and "test the waters" with us. That's because we want to be a part of your solution and that means more than just bringing computer hardware to the table. Folding is a great project and I wish the best for it's future.

[CBT]

... we have had a large company donating computer time anonymously ... That covers about ~30,000 CPUs or so. ... that ended around the new year, ....
I am hoping that this group will let us publicly acknowledge their contribution soon as what they've done (and the work we've been able to do on those machines) has been pretty exciting for us.

Would this organization be willing to publish on their reasons for joining the FAH project, their internal processes around FAH, the setting they ran FAH in, and their experiences, as input for other organizations that are looking into this? Also their reasons for stopping may be valuable information.

[Adak]

CBT, how can they publish this information, and remain anonymous as donors?

[bruce]

Sure they can, though they may not want to.

The following is total fiction. Any resemblance to the facts in this instance are purely coincidental.

We are a Fortune 500 company who wish to remain anonymous. We have more than XX,000 PCs used mostly for office applications during the hours of X AM and X PM spanning X time-zones. In order to examine the impact, if any, and feasibility of donating resources to FAH, we installed XX,000 clients with the following options: XX, XX, and XX. After X months of testing, we determined that XXXXXXXXXXXX. We also concluded that the value of of the research we donated to FAH was {more,less} than our costs and we plan to {continue, discontinue} donating.

[Bill1024]

I hope they also said "We installed separate electric meters on the systems folding so we don't deduct that electric cost from our taxes by mistake"
Since that electric is not tax deductible, we would not any trouble with the IRS. Sure they did.

[VijayPande]

Also their reasons for stopping may be valuable information.

In most cases, the donations we get from large corporations are for pre-set lengths of time. They 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.

I've made a blog post about this to give more details.
http://folding.stanford.edu/home/cpu-co ... tnerships/

[Leonardo]

I appreciate the information.

[Hukkel]

What a seriously bad decision to eventually END BA folding.

What are you guys smoking?

People have invested thousands and thousands of dollars and euros into getting big BA systems running. ll folding for the cure. People committing to folding, wanting to help out in the biggest way they can.

And now you guys say "hey man twas fun, we don't need you anymore. Sorry about those 10.000s of dollars you invested in us for nothing. BYE!!!"

Unbelievable. I cannot comprehend this. How can you guys be such megalomaniacs to just decide this?

If this will not be turned around I will turn my back onto folding altogether. Having spend so much money just to see it evaporate like this.

[ChristianVirtual]

You still have a full (!) year and after that can continue folding regular SMP projects. We all know that today that's not attractive but with the announced rework on the QRB formula I can imaging that the faster systems will get a higher PPD; give PG some time to work it out.

[Grandpa_01]

What a seriously bad decision to eventually END BA folding.

What are you guys smoking?

People have invested thousands and thousands of dollars and euros into getting big BA systems running. ll folding for the cure. People committing to folding, wanting to help out in the biggest way they can.

And now you guys say "hey man twas fun, we don't need you anymore. Sorry about those 10.000s of dollars you invested in us for nothing. BYE!!!"

Unbelievable. I cannot comprehend this. How can you guys be such megalomaniacs to just decide this?

If this will not be turned around I will turn my back onto folding altogether. Having spend so much money just to see it evaporate like this.

I talked to Vijay a couple of days before the decision, and as he explained here in this thread he had his reasons for stopping bigadv and although I do not agree with his decision and he knows / knew that before the decision was made I do support him and his decision, who am I to second guess his decisions as to what is better for the project and science. I also know that he does care about the donors and there investment and involvement in the project he expressed that through his comunications with me and through his actions and reactions to thing that were said. (I have no doubt about that) Bigadv has not been removed from the donors yet that is still a ways away.

While many invested in equipment that would do bigadv the original goal of the project was not being met which was the top end equipment and in reality the bigadv program was not set up properly, to begin with there should have been a road map clearly stating the goals and expectations of the class and tpf adjustments needed and updated on a regular basis. If that would have been done in the beginning most if not all the strife could have been avoided.

Vijay is aware that if there is not a great enough value for the bigadv class hardware to run it will most likely be shut down or moved to other projects or tasks, thus the new points system I do not know what he has in mind as far as qrb goes I did not ask.

I do know that many of the individuals that are running bigadv class equipment have already shut them down, I myself know personally of 20+ 4P machines that are no longer folding F@H on them that is over 1000 cores by itself. I would imagine that that is slowing down the bigadv project even more and that PG may be forced to send it to the cloud due to lack of donor participation. I know Vijay does not want to loose that type of donor folding power and would imagine he will get the new points system out as soon as possible there are some very discouraged donors out there right now who need a little encouragement just like yourself.

I believe F@H has the best protein folding project available, I believe it has the greatest chance of finding useful results that will benefit my children, grandchildren and future generations of man kind. I also know that if the reward is greater on some other type of equipment these rigs will be shut down or run on something else, and any upgrades will be to whatever is paying the best at the time. Vijay is not dumb he knows the equipment will follow the points that is what a points system is for (Points incentive), I am pretty sure that if he has a need and wants to keep the server type systems the points system will reflect that, because if a person can make more PPD/Watt and $$$ spent on GPU then that is where most of the donors will go.

He is working on the other problems that were brought up also which I believe were just as severe if not more so as the bigadv issue, Hopefully Vijay will keep us updated on a regular basis until he gets a PR person in place and the alienation of donors continues to be put on hold here at the FF. We shall just have to wait and see what happens and hopefully it does not fit into the joke category of (soon) because I do not believe that would be a good thing.