Running F@H and BOINC at the same time

[DynamicInsanity]

Hey,

as i'm mostly interested in cancer related projects i thought it might be good to let f@h run an my gpu und the world community project the help conquer cancer part on my cpu.

What i wanted to know is if it is better to let both projects run or concentrate on one ?!
Oh and for my Hardware i'm going to buy a new pc in the next few days so i cant give you more detailed information.

[bruce]

Welcome to the foldingforum, DynamicInsanity.

I'm not sure what kind of an answer you expect. You're here at the FoldingForum asking for an answer from people who have already decided that FAH is better than BOINC. That's a decision that you have to make for yourself.

Yes, it's possible to run both at the same time. Running more than one CPU-based client is not recommended because both of them want to use 100% of the CPU and juggling priories is really difficult.

Running a F@H GPU client and a BOINC CPU client is probably better use of your resources. All GPU clients need a certain amount of CPU resources to move data to/from the GPU even though the heavy work is being done by the GPU. When running F@H on a NVidia GPU, very little CPU is used; with the current software, an ATI GPU needs more CPU. (Most of us hope that the new software will help ATI in this regard, but nobody knows yet.) The F@H client has two choices of priority: IDLE (Lowest Possible) and LOW (Slightly Higher). Choose the latter, I'm not sure if BOINC has similar choices, but any application that uses large amounts of CPU resources should be at a lower priority than one that uses small amounts more frequently especially when it's just needed to deliver more work so the GPU stays busy.

By the way, some of the FAH research projects target cancer, too.

[DynamicInsanity]

Hey,

thx for your answer. Hmm what i wanted is if it's better to focuse my resources on FAH or WCG only. Well i don't know if its possible to get an answer at all but what would be the better choice to focus only on one of the projects or split my resources ?! Well from a scientific point of view. I hope it's easier to understand now what i mean.

My new pc i am trying to get is only for folding/cloud computing stuff ( i guess you know what i mean)
Just for you to know that my resources wont be very limited.

[bruce]

I am firmly convinced that focusing on FAH is better than focusing on WCG. I'm basing that on my understanding of the science being done. I don't know any better way to answer your question.

[hj47]

I'd just like to interject for a moment. I really don't want to start a BOINC vs FAH circlejerk thread, but I'm going to have to respectfully call you out on your assumptions, Bruce.
I wouldn't make such broad sweeping statements like

people who have already decided that FAH is better than BOINC.

. Well, no, I for one haven't reached that conclusion. And I'm sure many other FAH users haven't either. There are numerous biological projects on BOINC that deserve recognition but are currently undervalued and under-participated. GPUGrid for example. I keep my GTX 460 fed on GPUGrid units because the science that these guys are doing is very interesting. They are going down avenues that FAH does not, i.e. focusing on large scale polymers and superstructures other than proteins. They have written quality peer reviewed papers based on the results, and that is a good thing. Both projects are good for science, and are both complementary. In no way are they in competition.

Another project is rosetta@home, that, IMHO is producing more exciting results than FAH. But Rosetta doesn't aim to replace, or even compete with FAH; they are tackling the protein enigma from two very different perspectives, and from two different timeframes. Rosetta is more about identifying, testing and designing proteins in the already folded state, while FAH investigates the actual folding process. The former arguably has shorter term scientific feedback, while the latter really is a long term investigation. This is not to say that Rosetta is better than FAH, but complementary. And thats the thing with science, investigation needs to be broad and diversified. A single avenue of investigation is not enough.

That's why I choose to run all three DC projects. All three are contributing to the growing science of molecular dynamics and simulation. You cannot disagree with that. And I think you need to be more constructive with your criticism, you of all people Bruce.

tldr; BOINC projects are just as important as FAH, as science is not a one way street of investigation.

[Grandpa_01]

You can run them both at the same time. I do run BOINK on my ATI GPU's because they do not preform very well on FAH, if you try to fold with them at the same time as running smp you will slow down the smp WU significantly. I can run Boink and fold smp at the same time with very little loss to the SMP. If and when the new GPU folding client comes out and does not penalize me for running FAH on it I will switch back to folding with my GPU's. But until then I will donate my GPU use to Boink projects. I feel it is better to use them for something useful than just sitting idle while the CPU's fold. But as far as I am concerned FAH is the better choice it is managed better and has dedicated staff of professionals who are very involved with the process. And that makes me feel that they care about the outcome and what they are doing.

[[email protected]]

Wow, This was very informative. Very good read and thought provoking about the actual science.
Thanks.....

[WoodburyMan]

I always go back to BOINC with SETI@Home every few months for a day or two and always get frustrated when i do. The little things bother me with BOINC. Seti@Home's BOINc servers are _always_ down or don't have work.. and they do not have a true multicore capabilitys for SETI@Home project yet.. and other small things. What it boils down to is that Folding is doing real science. While finding ET is nice, finding a cure for cancer or other diseases is more worth my time IMHO.

[hj47]

I always go back to BOINC with SETI@Home every few months for a day or two and always get frustrated when i do. The little things bother me with BOINC. Seti@Home's BOINc servers are _always_ down or don't have work.. and they do not have a true multicore capabilitys for SETI@Home project yet.. and other small things. What it boils down to is that Folding is doing real science. While finding ET is nice, finding a cure for cancer or other diseases is more worth my time IMHO.

You do know that there is more than 1 project running on the BOINC infrastructure, right?

[jaxawier]

Think both f@h and boinc complement each other,had a time when I tried several boinc projects,some of them had those issues mentioned with no work,or very little of it,days for a reply on the forums if ever,that goes for the smaller projects.Bigger projects like milkyway or gpugrid doesn´t have that issue,both is nice for gpu´s, but over at the boinc-pages every project is listed,and for graphics they have a spearate section for it,if it is ati/amd or nvidia or both,so that couldn´t be easier imho When that is said milkyway was a resourcehog,no problems if it is dedicated but I might run some other apps on that pc,a nightmare to vnc into it and everything hangs,and gpugrid have several clients or workgroups,some of them crashed for no reason even on stock,for gpugrid they have advices on their forum to solve much of that so that worked perfectly in the end,one occasional crash but f@h crashes too from time to time

[bruce]

At this forum, it's natural to find folks who feel F@H is superior to BOINC. We don't promote BOINC, but there are a number of good projects that use the BOINC infrastructure so we don't bad-mouth it either.

What's really important is that you manage your resources carefully.

Turn-around time in FAH is critical, and unlike BOINC, it's not sufficient to just meet the deadline. For that reason, we recommend against sharing resources between FAH and BOINC. Even though you may make it work, FAH would be better off if you ran one or the other or you decided exactly which of your resources you're willing to DEDICATE to FAH.

Each FAH Uniprocessor client needs a dedicated CPU (not a HT virtual processor or a BullDozer core that shares a FPU with another task). The FAH SMP client needs all of your CPUs to be dedicated to it. The FAH GPU client is a bit strange because it's performance depends both on which type of GPU you have and how much CPU is available to it. An ATI GPU will use most of a single CPU, though it can be a virtual cpu). A Fermi GPU will use quite a bit of a CPU. Pre-fermi NV GPUs need very little CPU time.

As long as FAH can have dedicated resources or you can manage CPU priorities so that BOINC only uses the CPUs when FAH has nothing to process, we welcome you to run both. If not, I recommend you run BOINC one week and FAH the next week, carefully using the -oneunit flag (or the Pause When Done menu option) to finish the current WU before shutting FAH down for more than a short period.

[JimF]

It depends a lot on your hardware, with some combinations being more feasible than others. I run FAH on two PCs with a total of five Nvidia GPUs (four GT 240s and one GTS 450), because they have high PPD/watt output and use essentially zero CPU resources (even the Fermi card is now usually about zero percent CPU with the recent drivers, though that depends somewhat on the project being folded). That leaves me a dual-core and quad-core to run BOINC/World Community Grid. I let WCG select the projects for me, since IBM and its advisory group knows more about the science than I do, and they keep their servers running reliably. And it works great on my PCs, with no interference between the projects at all. I do occasionally get a little screen lag from folding, but can deal with that in various was discussed on these forums, including pausing the folding if necessary, and it does not appear to be related to the BOINC use of my CPU. And it will give you something to do keeping up with the software upgrades for two projects rather than one.

And note that I use Intel CPUs, since the software seems to be more optimized for them, but I don't know much about AMD.

[Rattledagger]

At this forum, it's natural to find folks who feel F@H is superior to BOINC. We don't promote BOINC, but there are a number of good projects that use the BOINC infrastructure so we don't bad-mouth it either.

What's really important is that you manage your resources carefully.

Turn-around time in FAH is critical, and unlike BOINC, it's not sufficient to just meet the deadline.

While turnaround-times doesn't matter to some of the projects, like SETI@home, it does matter to others, and example GPUGRID, just like Folding@home, therefore gives a fast-turnaround-time-bonus...

For that reason, we recommend against sharing resources between FAH and BOINC. Even though you may make it work, FAH would be better off if you ran one or the other or you decided exactly which of your resources you're willing to DEDICATE to FAH.

Each FAH Uniprocessor client needs a dedicated CPU (not a HT virtual processor or a BullDozer core that shares a FPU with another task). The FAH SMP client needs all of your CPUs to be dedicated to it. The FAH GPU client is a bit strange because it's performance depends both on which type of GPU you have and how much CPU is available to it. An ATI GPU will use most of a single CPU, though it can be a virtual cpu). A Fermi GPU will use quite a bit of a CPU. Pre-fermi NV GPUs need very little CPU time.

As long as FAH can have dedicated resources or you can manage CPU priorities so that BOINC only uses the CPUs when FAH has nothing to process, we welcome you to run both. If not, I recommend you run BOINC one week and FAH the next week, carefully using the -oneunit flag (or the Pause When Done menu option) to finish the current WU before shutting FAH down for more than a short period.

Running FAH's Nvidia, Ati or single-cpu-client alongside BOINC is easy, it's just to decrease #BOINC-cores accordingly (if cpu-client). It's also easy to run FAH's SMP-client alongside one (or more) BOINC GPU-projects.

But, atleast my opinion is the "pay" is too low on the FAH-single-cpu-client compared to the SMP-client to be worth it, assuming fast enough computer to run SMP-client that is, so I'll recommend to run the SMP-client and BOINC together this way:

FAH-side, a small batch-file with the cmd-client, something like this (*):

Code: Select all

:START
call fah-smp-client.exe -oneunit
call sleep 86400
goto START

In case doesn't already have it, you'll need to download a Sleep-application. The 86400 is how many seconds wait until should start FAH again, it equals 24 hours, so to increase/decrease the wait just increase/decrease this number.

BOINC-side, a cc_config.xml located in BOINC's data-directory, minimum including:

Code: Select all

<cc_config>
<options>
   <exclusive_app>FahCore_a3.exe</exclusive_app>
</options>
</cc_config>

Then neccessary, just add more lines with <exclusive_app> then other FAH-cores would be used.
There's also a <exclusive_gpu_app> that can be used in case would run both a FAH-gpu-client and a BOINC-GPU-project (except GPUGRID that is).

The advantages of running things this way, compared to running "1 week FAH, 1 week BOINC", is that BOINC will immediately take-over in case problems getting work from FAH, won't sit idle waiting on uploads/downloads and no idle time switching between clients. Depending on BOINC-project, in some BOINC-projects 7 days is the deadline so crunching one week FAH would either mean some wu's crunching-time would be wasted, or, cpu-cores is part of the time idle waiting until all BOINC-work is finished before can startup FAH again. Exiting client instead of pausing will also waste anything crunched since last checkpoint, depending on project this can be everything from 1-minute between checkpoints upto multiple hours between checkpoints, so a shutdown at the wrong point can potentially waste many hours crunching.

So for anyone wanting to run both FAH and BOINC on cpu's, by using BOINC's <exclusive_app>-mechanism you'll get the same turnaround-time running FAH as if you're not running BOINC at all, you'll maximise FAH-points/day by running SMP-client instead of single-cpu-client(s), and you'll not sit idle for various reasons.


(*) Not tested with v7-client.