New Project - QuChemPedIA@home

[Dataman]

I do not know much about this one as it is all in French (my French is limited to "merci" and a few curse words  ). It requires an invitation code and VBox64. The site is up but only a few (56 when I looked) wu's.

Google translate: This project is in the fields of data sciences and artificial intelligence. The goal is to revolutionize the field of quantum chemistry and molecular computing.
Molecular chemistry is defined as the study of discrete entities (molecules) and corresponds to the widest community of chemists. Hundreds of millions of molecules are known, usually containing less than a hundred atoms and less than a thousand electrons.

The chemical properties of these molecules depend on the position of the electrons, which can be calculated in an approximate way using methods from quantum mechanics. The various calculation parameters define the quality of the method and its cost in time (from a few hours to a few thousand hours per molecule). With the democratization of computational power, computer chemistry has become an essential part of chemistry research. Today, it is impossible for a human to take into account the results, even limited to the most important data, for millions of known molecules. For the chemists, the ambition of this project is to radically change the approach using artificial intelligence and data mining methods in order to more effectively explore the immense combinatorial space of the molecular space.

The project is therefore at the frontier between artificial learning when it comes to predicting the properties of a molecule and combinatorial optimization when we seek to propose new molecules that maximize one or more objectives. The proposed methodology, which combines deep learning and constraint programming to generate molecules, is new to our knowledge and offers remarkable opportunities. Associating a cost function while predicting the synthesis routes makes it possible, for example, to propose molecules that are less costly for the environment, the basis of the green chemistry of tomorrow. The search for new photovoltaic materials is also the first of the applications we are considering.


https://quchempedia.univ-angers.fr/athome/


Sounds interesting so I am going to try and figure out how to request an invitation code.





EDIT: The message board is empty so I tried a generic email to mail@<domain name> and courrier@<domain name> but they bounced back as return mail. Looked up the project at BOINC Berkeley and there is some discussion there but no invitation code.

[Dingo]

I tried to send an email at admin@quchempedia.univ-angers.fr but it bounced back.  Have to wait and see I guess.

[Daniel]

I joined this project today.
They now have an invitation code on their website.
They also have a project description in English and the message boards are working.

They have made an announcement that the project is open ...

Public opening
We are pleased to announce the official opening of the quchempedia@home project.
Thank you for your precious help !

I have downloaded a few workunits, but none of them have finished yet (they have been running for about 3 hours so far).

At the moment I don't see the stats for this project on the BOINCStats website, but the project admin has given permission for the stats to be displayed, so I guess it won't be too long before they are added.

[Dingo]

The Team was created and it is open to new members.  I attached but there is no work at the moment.  It is a combination of VBox and standard work.  I will make a forum page tomorrow.


Microsoft Windows running on an AMD x86_64 or Intel EM64T CPU   0.08 (vbox64_t1)   3 Oct 2019, 20:59:44 UTC   0 GigaFLOPS
Linux running on an AMD x86_64 or Intel EM64T CPU   0.08 (intel_mt) (beta test)   3 Oct 2019, 20:59:56 UTC   0 GigaFLOPS
Linux running on an AMD x86_64 or Intel EM64T CPU   0.08 (t1)   3 Oct 2019, 20:59:54 UTC   0 GigaFLOPS
Linux running on an AMD x86_64 or Intel EM64T CPU   0.08 (t2)   3 Oct 2019, 20:59:57 UTC   0 GigaFLOPS
Intel 64-bit Mac OS 10.5 or later   0.08 (vbox64_t1)   3 Oct 2019, 20:59:46 UTC   0 GigaFLOPS
Total average computing: 0 GigaFLOPS

[tazzduke]

Greetings All

I have joined and managed to get 1 workunit, I am also doing some Universe@home

Cheers

[Daniel]

These workunits are taking a long time.
So far only one of mine has completed.
It took about 8 hours, but all the others are still running.
My longest running one has been going for about 19 hours.
It looks like they are still doing something - my CPU usage is at 100%

[Dingo]

I ran a bunch of Linux tasks as they are all CPU.  The shortest was 3271 seconds and ended up at 116.29 credits per hour.  The longest was 66423.81 sec and got 18.45 credits per hour so a big difference and I have no idea what they did ?  I still have one running for 21:25 hours and it is estimated to complete in 13:35 hours

[NudgeyNR]

Just joined and attached on windows 10 and am running NWChem 0.08 (vbox64_t1) and remaining time is nearly 9days(est), using nearly 97% CPU (per core)

[Daniel]

Yes, they take a long time.

So far 4 of my workunits have finished ...

Run time (sec)    Credit
28,818.60           297.23
91,158.23           940.19
14,003.25           144.43
82,897.88           854.99

The others are still going.
My longest running one is now at 1 Day 14 Hours.
I haven't had any errors or invalid ones yet.

[Dataman]

I joined too. I limited it for now to one wu and one cpu until I see how it runs. Wow, what a flame thrower. Like Nudgey, it is using 98% of the core and the run time is almost 9 days. 
I will see how it goes before committing more resources. Looks like an interesting project.

[kashi]

OK, downloaded one too.

Haha, task has 8 days 23 hours left to run. Come on, giddy up NWChem, put on your dancing shoes, let's rock.

[Dataman]

I gave up. After 24 hrs only 2.1% of the wu had been processed. They have a lot of work to do before this project is acceptable.

[kashi]

Oops, busy with other things and didn't check. Nine tasks have all errored over the last 18 hours or so. VBox projects are just yuk really. I'm outta here.

[Daniel]

3 more of mine have finished ...

Run time (sec)    Credit
127,646.86        1,316.53
151,448.44        1,562.01
156,365.36        1,612.73

4 more are still running.

I think the percent complete and time remaining are not accurate.

So far mine have been getting to 100% with no time remaining, but they continue running for a long time after that.

The admin made a comment about it on the message boards ...

Runtime is not predictable (not determinist), perhaps the prediction of the scheduler is totally erroneous.

My longest running one is now up to 2 days and 8 hours.

I'll just leave them and wait and see what happens.

[Daniel]

Stats for this project are on the BOINCStats website now.
BOINC@Australia is currently ranked 8th in the world and is the only Australian team with credit.