Xansons

[Bryan]

Xansons will run out of work in the next 3-5 days. SUSA passed both SG and OcUK yesterday to take over the #1 spot. We had a 1.2M lead on SG but overnight they caught up to us.

somanyroads will get to the 100M milestone tomorrow (hopefully). He and I will be jumping on Dave GPU's account and try to get the 3rd SUSA member to the 100M level. I'm not sure there will be enough work available to get him there but we are going to give it a try.

If you would like to jump in with either GPUs or CPUs (pays well on CPU) and help Dave then please contact him. If you have your own goals then you need to jump on it immediately, if not sooner, and get things done.

Hopefully SUSA can be on top of the pile when the project completes!

[Mumps]

I've been running it on CPU for a while to see how far I could take it, but am not getting enough work to keep things busy. Hopefully you guys can hammer it enough to stay ahead of SG to the end. I'll add it to my list of projects my hosts keep hitting Update for though, because the less work SG has available to grab, the better. :).

[Bryan]

They were having trouble supplying enough CPU work until about 5 hours ago. All night my CPUs were at about 25% filled.

SG has overtaken us. They had several new people sign up on yesterday's update and there are 4 new ones today. I guess they are taking this seriously :( The pisser is that it is taking WU that are needed to meet other goals. I'd really like to see Dave break 100M but we are running out of time/work.

[Mumps]

Well, according to Free-DC, we moved back ahead of SG, but according to current stats on the project, SG is back in front again. If we want to hold it, we'd better get working a little harder.

[Bryan]

It is doing much better on supplying work tonight. It was a major pain in the backside all day. Uploads/downloads took multiple tries etc etc. Now it is running very smoothly.

An hour ago somanyroads and I have moved over to Dave GPUs account. I'm not optimistic that Scole, somanyroads, Dave, and I have the horsepower to get him over the 100M mark. Only time will tell and there isn't much of that left.

[Haunted Forest]

Looks like I am downloading a lot!
see it it runs:D

[FourOh]

I can't seem to keep all my machines running the project. Over the last few weeks I've been travelling a lot for work, if I'm not watching I'll get hung up uploads and eventually BOINC stops asking Xansons for more work. I think I have everything up now... maybe I'll get to 50M but doesn't look promising. We just need to stay ahead of SG!

[Bryan]

@FourOh ... here is the information for a script file that will go out to each of your machines every 2 minutes and tell it to "retry" the pending network communications. Just follow the instructions and then launch it. This is copied from a message I sent the guys on Dave's account.

The project is having problems with network communications right now and uploads/downloads are going into backoff mode.

I'm using a script (stolen from Scole during the Pentathlon) to tickle each of my machines every 2 minutes and retry any pending network communications. It looks like this:

:top

"C:\Program Files\BOINC\boinccmd" --host 10.0.0.31 --passwd xyxyxy --network_available
timeout /T 40
"C:\Program Files\BOINC\boinccmd" --host 10.0.0.51 --passwd xyxyxy --network_available
timeout /T 40
"C:\Program Files\BOINC\boinccmd" --host 10.0.0.50 --passwd xyxyxy --network_available
timeout /T 40

goto top

To use this then:
1. change the host IP addresses to match your machines
2. if you have more than 3 machines then add more entries to cover your hosts.
3. change the "xyxyxy" after passwd to your remote password.
4. change the wait time (T 40) to what you need. I'm hitting it every 2 minutes (120s). Divide 120 by the number of hosts that you have and that
is the number that goes in place of the "40".
5. if you did a standard install then it should work as is. If you did a custom install then change it so that it points to where your boinccmd file is located.
6. save it as a batch file (.bat extension).
7. double click the .bat file to launch it.

Every 2 minutes it will hit each machine and tell it to retry any pending network communications. It beats the hell out of hitting "retry" multiple times

[Bryan]

Xansons is finished with their primary work. They are rerunning some of the very 1st WU - the admin said about 10 hours worth.

1. We are about 4.7M short of getting Dave GPU over the 100M hump. If anyone has any spare GPU power please contact myself, somanyroads, scole, or Dave GPU for his login info.

2. If you are/were running for your own goals please DO NOT STOP until all work is gone. Since the CPU work went away (ie Mumps went silent) SG has cut into our #1 team lead by 800k. Obviously they had more GPUs on it than we did. Our lead is now 4.2M but I don't know how much the remaining work is worth. We might be safe and we might not.

[Mumps]

Woot! I see DaveGPU made it over the 100 Mil Mark! Congrats to all of you folks that made it there before the "end!" :)!

[Bryan]

And SUSA finished as the top team ... and is still adding to that lead =))

[Mumps]

Work is available again! Get it while we can!


3032 new structures were added for processing
A bunch of new COD structures (for October and November 2017) was added to the project. From now on the updates will be more regular (about twice a month).

Also, now the structure is considered processed by the user if the user calculated at least one diffraction pattern (x-ray or the neutron one, does not matter) for this structure. So, the numbers of processed COD entries are increased for all users. Now, it is also possible to view the complete list of the COD structures processed by the user (see 'view all' link near the number of processed COD entries). Note that displaying this list may take some time for the top participants who have 100,000+ structures processed.

26 Nov 2017, 21:44:59 UTC