Thanks for the update, it was needed.

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

If I may say so, POEM works well as a backup project on Maxwell cards. They are even more efficient than the AMD cards that I have checked, even though the current POEM application was developed using OpenCl for GCN architecture.

So set your resource share to "1" for POEM, and keep it at "100" for GPUGrid, and have the best of both worlds if you have the right card and want to do protein/biomedical research. For astronomy, there is also Einstein that works well with Nvidia cards too of course.

At this juncture (Scientist production work is limited): would turning on the optional ACEMD beta (test) application be industrious for GPUGRID ?

Thank you Gerard for filling up both queues.

Ah yes -- but at the moment, the POEM site is unaccessible....

Try this project instead:

http://www.einsteinathome.org/

Einstein @ Home GPU Work units finish relatively quickly, so if you set its resource share to 0, you'll get GPU grid work as soon as it's available.

Yes - I know -- I've got a lot of projects active <smile>


____________

Is it safe to assume there will be no new tasks for a while?

I wonder if other projects are suddenly noticing a boost to their throughput...

I'm running Einstein on priority zero so that I'll be able to pick up GPUGrid work once it's flowing again.

Thank you Gerard for the new workunits, and I wish you Merry Christmas!

I'm running Einstein on priority zero so that I'll be able to pick up GPUGrid work once it's flowing again.

I'm running Einstein on priority zero so that I'll be able to pick up GPUGrid work once it's flowing again.

could anyone please tell me how I can set a given project (e.g. Einstein) to "priority Zero"?

I, too, have Einstein (and others) running right now as a temporary replacement for GPUGRID, until there are new tasks coming.
At any rate, I want to priorize GPUGRID.

Go into the project preferences on your account page and set the project's Resource Share to zero. Once this is done, they will only request work if there is no work for any projects with a non-zero Resource Share. Once this is done, your computer will continue to receive work from whichever project(s) you have going but will continuously check for new work at GPUGrid if you have its Resource Share set to anything above zero and will only work on those other projects if there is no work available for GPUGrid.

Gerard filled the long queue couple of times in the past days, so my hosts (I think other's hosts too) had enough work.
Thanks for his enthusiasm!

I now could set Einstein and MooWrapper to Zero Ressource share. However, no way to get into the same screen for GPUGRID in order to set this to non-Zero. Please let me know how I can do this.

Look for the "Preferences for this project" and click on the "GPUGrid settings" -> ...

Look for the "Preferences for this project" and click on the "GPUGrid settings" -> ...

many thanks, now I got it. The value was at "100" anyway, so I didn't need do change it.
However, still no GPUGRID tasks coming in. Is the recently announced shortage of new tasks still continuing?

Thank you Gerard for filling up the queue again.
I wish to all of us a happy new year, full of short & long workunits! :)

also from me best New Year greetings to everybody :-)

Let's continue much crunching in 2016 !!!

I too wish a Happy and Crunchy :D New Year to all GPUGRID volunteers and scientists!
____________

hello everybody - just a question, to be on the safe side that nothing is wrong with my settings: I have not received any WUs since yesterday.
Are there none available for the time being?

GPUGRID has been low on WUs for some time now. You can check the Server status page to see the current availability of WUs.
____________

GPUGRID has been low on WUs for some time now. You can check the Server status page to see the current availability of WUs.

There are many users who help the GPUGrid project. Please stop complaining about new tasks being unavailable.

A good approach would be to simply setup backup projects, with Resource Share set to 0, to keep your GPUs busy, until BOINC happens to request GPUGrid tasks at a time when they are available.

Another good approach is, if you have projects that you want to do CPU work for, but only want to do GPU work for only if GPUGrid is out of work ... then set GPUGrid to a ridiculously high Resource Share, and set those other projects to a sufficiently low Resource Share. For instance, you could set GPUGrid to 10,000, and leave those other projects at 100.

Some example projects you can use as backups are: Poem@Home, SETI@Home, Einstein@Home, Asteroids@Home.

We understand that you want to help GPUGrid as a priority. We all want to do that! But, instead of complaining, the best thing you can do is to let BOINC do its job, let GPUGrid admins do their job, and setup backup projects to keep your resources from going idle.

Regards,
Jacob

I am not complaining, but observing. We do go through these WUs fast. One day we have hundreds, the next day there is nothing!

Hi,
somebody might have noticed that we recently running low in WUs.

The main reason for this is that we have very few people at the moment. Practically only Gerard is using it. Noelia is finishing her Phd in February and mainly writing up. Nate finished in July.

The good news that we are hiring more PhD students and already from January the group should be quite a bit larger with at least two other phD students and possibly one postdoc.

GDF

So, when will the new PhD students be on board?

Since I am planning, for the very near future, a major investment of several thousand Euro into additional new hardware just for GPUGRID crunching (indeed, no kidding!), it would, of course, be kind of important for me to know what's going on at GPUGRID.

Perhaps someone can enlighten me and all of us out here :-)

Since I am planning, for the very near future, a major investment of several thousand Euro into additional new hardware just for GPUGRID crunching (indeed, no kidding!), it would, of course, be kind of important for me to know what's going on at GPUGRID.

Perhaps someone can enlighten me and all of us out here :-)

I think your several thousand euro would be much better put on a donation to GPUGRID, if you really want to advance the research it performs. It could be used, for example, to finance a PhD student.

I think your several thousand euro would be much better put on a donation to GPUGRID, if you really want to advance the research it performs. It could be used, for example, to finance a PhD student.

The work we do here in crunching the data is apparently just about useless. According to Mr Giannadakis, at any rate.....

????

Probably the real limitation is time. They are on an academic year cycle. People come and go. This is not like an industrial research operation where they have a large permanent staff. And if it were, they most likely would have their own supercomputers and would not need us.

I'm getting no new work units now for two weeks???? Any ideas why? No changes on my end and I'm supporting three other projects with no problems. My CPU is just sitting there waiting?

Click "Server Status" at the top of the page.
Look at the "Unsent" column.
If they are all 0's, then there's no work available.

Simple as that.

I recommend adding other GPU projects, to keep your GPUs busy - and if you want them as backup projects, set their Resource Share to 0.

My CPU is just sitting there waiting?