Hi everyone,

yesterday we launched a bit more than 13,000 short WUs called *CXCL12_chalcone_umbrella*. They are pretty small WU, of 8ns (compared to a normal long WU ~40ns) and we hope they are easy and fun for you to crunch. Please post any problem you may encounter.

Scientists have been using a technique called Umbrella Sampling for some time now, with relative success in determining what we call binding free energy (which indicates how strongly a drug can bind its protein target). It is a pretty straightforward and much less expensive technique compared to the one we regularly use in our lab (adaptive sampling). However, it is particularly error-prone if the assumptions we take are wrong.

We are particularly excited about these WU, because while most scientific effort have been focused on reproducing free energies for single particular models or (most of the time) very simple toy models, it is the first time to our knowledge that, thanks to the fantastic community we have built together in GPUGRID, we can use this technique in a real case to screen potential drugs binding to CXCL12, a chemokine related to cancer metastasis.

Thanks to everyone for your contribution and I will be happy to assist you if you find any problem on the way. :)

Hi Gerard,

These workuints, just as the "GERARD_A2AR_NUL1D" (long) workunits are running with very low GPU load (32~53%), especially when the CPU is crunching too. Can you add any comments on this?

Thanks for the heads-up Gerard! :)
____________
[CSF] Thomas H.V. Dupont
Founder of the team CRUNCHERS SANS FRONTIERES 2.0
www.crunchersansfrontieres

Hi Retvari,

I've been reading a lot the same question in the forums so I think you deserve a concrete answer. The workunits that you refer to include "extra" forces that, because of the simulator implementation, cannot be calculated with the GPU and must be calculated via CPU. I guess that the communication between CPUs and GPUs for each of the simulation step is definetely a bottleneck causing a decrease in GPU performance.

I've made some figures for the curious ones, explaning these "extra forces".

*A2AR* batch




These are membrane systems. You can see the protein in yellow, embedded in a membrane (sticks) and solvated in a water box (blue cage). In these models, we place a drug (marked as "ligand") in the extracellular space, where it has to bind with its receptor (the protein).
Because of the way we simulate, these systems have a property we call "periodic boundary conditions", meaning that each of the molecules on the sides of the box, interact and can actually flip to the opposite side of the box. This allows the nice effect that the system is solvated in a "non-finite" box. Because of this effect, the ligand can jump from one side of the box to the other one. However, this doesn't make any biological sense, because drugs placed in the extracellular space can't freely access the intracellular space (they can't jump the membrane!). To ensure that the ligand stays in the extracellular phase, for each simulation frame we calculate the position of the ligand and if it is higher than the red line we apply a force down to make it stay. This arbitrary "force" is considered extra and must be calculated via CPU.

*umbrella* batch



In this case, we use a technique called "Umbrella Sampling" to calculate the binding free energy of the ligand (in yellow) to the protein (in blue). This technique consists in assuming that the unbinding (or the binding) occurs in a linear way (green line), and what we do is to simulate the ligand in different positions along this pathway (ideally every 0.5 angstroms). After the simulation, we calculate how stable the ligand was in each of the positions and using some mathematical framework called WHAM we can calculate the probabilities that the ligand goes from unbound to bound. Now comes the "extra force": in order to force the ligand to sample the desired position along the pathway, we apply a force (in red) to make it stay there. This force increases as the ligand goes further away from the initial position, giving this potential profile of an inverse umbrella (in black). This is why is called "Umbrella sampling". Again, this "extra force" must be calculated via CPU...

I hope you found it useful or interesting! If you have any questions please do not hesitate. :)

EDIT
I have revised the A2AR_1D batch because it was indeed looking much slower than the other A2AR. I've made a change that should speed up new 1D WU you may get. I'll be sending *1Dx* batch soon, tell me if is faster for you guys.

Thanks for the explanation and time Gerard it is what makes me want to crunch GPUGrid unlike other projects who explain nothing.

Great news! Thank you!

Very good explanation Gerard, now we know what we crunch.
Thank you.
____________
Greetings from TJ

Thank you, Gerard

Thanks for your time Gerard! Really appreciated! :)
____________
[CSF] Thomas H.V. Dupont
Founder of the team CRUNCHERS SANS FRONTIERES 2.0
www.crunchersansfrontieres

My statistics are still rather thin, but with one GERARD_CXCL12_chalcone_umbrella completed on each of four Maxwell cards:

GTX 960: GPU Load 34% (supported by 1 core of i7-4790)
MCL: 16%
Power: 33.3% TDP = 40 watts
Time: 4 hours 11 minutes (ave between the two cards for 2 work units)

GTX 750 Ti: GPU Load 49% (supported by 1 core of i7-4770)
MCL: 10%
Power: 31.3 TDP = 19 watts
Time: 5 hours 6 minutes (ave between the two cards for 2 work units)

None of the cards were overclocked by me, and only minimally factory-overclocked. They were so lightly loaded that they usually ran at less than their maximum GPU clock settings.

So it seems that the GTX 750 Ti is more efficient, running at 80% of the speed of the GTX 960, but using only 50% of the power.

Thanks for the explanation Gerard!

This one is failing in all hosts at uploading after completing correctly with :

<error_code>-131 (file size too big)</error_code>

https://www.gpugrid.net/workunit.php?wuid=11490316

EDIT
I have revised the A2AR_1D batch because it was indeed looking much slower than the other A2AR. I've made a change that should speed up new 1D WU you may get. I'll be sending *1Dx* batch soon, tell me if is faster for you guys.

We've changed the size limit for these WU. I hope this fixes this problem in the new WU. Sorry for the inconvinience!

We've changed the size limit for these WU. I hope this fixes this problem in the new WU. Sorry for the inconvinience!

Correction. it should be like this:

For these low GPU usage and high CPU usage WUs, if you use this:

<app>
<name>acemdshort</name>
<gpu_versions>
<gpu_usage>.5</gpu_usage>
<cpu_usage>1</cpu_usage>
</gpu_versions>
</app>


in your app_config.xml file, you can increase GPU usage from 30%-40% range to 60% to 70% range, depending on your hardware and OS.

That's what is happening on my computers.

Also note that this app_config.xml should be placed in the project's folder under the BOINC folder.
For example on Windows Vista, 7, 8, 8.1, 10:

The website access is also slow at times.


I hope these WUs do not become the "standard". That would be a total waste of high end GPU video cards.

I'm getting now the "transient upload error" and "server is out of disk space" messages in three units trying unsuccessfully to upload.

I'm getting now the "transient upload error" and "server is out of disk space" messages in three units trying unsuccessfully to upload.

same problem with my 4 hosts: no upload of finished WUs possible, no download of new WUs for crunching.

Server breakdown at GPUGRID ?

It means the server is out of disk space, again!

How could anyone on here and not involved with the servers possibly answer that question?

... I hope these WUs do not become the "standard". That would be a total waste of high end GPU video cards.

How could anyone on here and not involved with the servers possibly answer that question?

How could anyone on here and not involved with the servers possibly answer that question?

is it impossible that anyone involved with the servers is also reading in the Forum?

... they would stick another floppy disk in.

... they would stick another floppy disk in.

yes, indeed :-)

Yep, I have 3 GPUs, and all 6 of my tasks are completed with uploads stalled. I suppose this is a good test of GPU backup projects (they are working on Asteroids, Einstein, and SETI)... but I hope the GPUGrid admins get their upload/space issues resolved :)

Think I will give this a rest till long wu are back.

... they would stick another floppy disk in.

yes, indeed :-)

Perhaps we should donate a 8TB HDD for the project. Who's in?

... they would stick another floppy disk in.

yes, indeed :-)

Perhaps we should donate a 8TB HDD for the project. Who's in?

... they would stick another floppy disk in.

yes, indeed :-)

Perhaps we should donate a 8TB HDD for the project. Who's in?

... they would stick another floppy disk in.

yes, indeed :-)

Perhaps we should donate a 8TB HDD for the project. Who's in?

First, I'd like to know how much space job records like these are eating.

22 Jul 2014 invalid: https://www.gpugrid.net/workunit.php?wuid=9910195
5 Mar 2015 invalid: https://www.gpugrid.net/workunit.php?wuid=10721767

I think these are the work units to run on mobile notebooks. As notebooks are thin devices most doesn't have enough air ventilation to cool them enough on loads such as GPUgrid works.

However, my notebook with GT740M runs at about %85 - 90 GPU usage with these units and at 65C degrees with 3 units of WCG works running on the CPU. This is really colder than other units which reach 80 or more on the same situation. :)

The team is aware of the problem. They're working on it.

I have been getting the server is out of disk space for the past 24-hours.
____________

Should be fixed by now. Please let us know if you are still having this error!

Yep, I have 3 GPUs, and all 6 of my tasks are completed with uploads stalled. I suppose this is a good test of GPU backup projects (they are working on Asteroids, Einstein, and SETI)... but I hope the GPUGrid admins get their upload/space issues resolved :)

Also, POEM@home and Milkyway@home

Emailed Gianni regarding the disks full/upload problem.

no more "Umbrella" WUs coming ?

no more "Umbrella" WUs coming ?

Oh, I dunno: I processed 28 of these WUs successfully!

no more "Umbrella" WUs coming ?

I hope not. They were a pain!

All (71) · In progress (2) · Pending (0) · Valid (21) · Invalid (1) · Error (47) ....47vs71.. im crunch now only on old gtx 680 and laptop gtx 960 4gb.
im dont check long time my stats and message board here, just im now say wow :-)) hope they fix it soon for new gf..-)

141 wus crunched, only few errors at the beginning due to incorrect upload figure in the WU, then smooth.

Sure, I have a good internet connection.

I like these new GERARD_CXCL12_BestUmbrella WUs. They have good GPU usage and the output files were not too big. I successfully completed 3 so far.

Good work!

Can I have some more of them?

I am sorry for the Umbrella runs, they were somewhat experimental. However, the results look promising so far.

I can try to tune some parameters for future releases like the file size (we may be able to reduce it without big impact on the analysis results) but unfortunately they will still be quite CPU consuming because of the aforementioned reasons. Could you post the main reasons why they were a "pain"? :) We can try to find a common solution.

I am now sending plenty of classic long WU that should push your GPUs to the limit! :) However, I plan to launch more Umbrella short WU in the future, which to me seem ideal for old GPUs.

I'll keep you posted!

Could you post the main reasons why they were a "pain"? :) We can try to find a common solution.

I am sorry for the Umbrella runs, they were somewhat experimental. However, the results look promising so far.

I can try to tune some parameters for future releases like the file size (we may be able to reduce it without big impact on the analysis results) but unfortunately they will still be quite CPU consuming because of the aforementioned reasons. Could you post the main reasons why they were a "pain"? :) We can try to find a common solution.

I am now sending plenty of classic long WU that should push your GPUs to the limit! :) However, I plan to launch more Umbrella short WU in the future, which to me seem ideal for old GPUs.

I'll keep you posted!

[quote]need of 3 short task per GPU in the Queue (presently the limit is 2 per GPU)

[quote]need of 3 short task per GPU in the Queue (presently the limit is 2 per GPU)

once these WUs are distributed again, it really would make a lot of sense to raise the limit from 2 to 3 per GPU !

To offer a minority view, I was quite happy to get them. I don't normally run the shorts, but the science looks very interesting, and my GTX 750 Tis are not that good on the longs anymore. If they don't use much GPU power, but more CPU power, that is OK if that is what the calculations call for. You can't change the science or math just to heat up the cards more. Also, I have large enough upload bandwidth (4 Mbps) that I did not notice any problems there, or with memory, etc. But a warning as to all of this would undoubtedly be a good idea, since it may push many machines over the edge in one way or another..

However as Gerard has said in another post he is already overun with work and probably does not have the time to do any of these things.

High CPU usage -> low GPU usage -> need of 2 simultaneous short task per GPU -> need of 3 short task per GPU in the queue
(presently the limit is 2 per GPU)

the current short runs "Enamine_Umbrella" use some 50-60% of a high-end GPU.
As already said above by one of our power crunchers: the limit of 2 such WUs per GPU should be increased to 3, if not to 4.

the current short runs "Enamine_Umbrella" use some 50-60% of a high-end GPU.
As already said above by one of our power crunchers: the limit of 2 such WUs per GPU should be increased to 3, if not to 4.

any news on this?