rpi_logo
New MilkyWay@Home Developer
New MilkyWay@Home Developer
log in

Advanced search

Message boards : News : New MilkyWay@Home Developer

Author Message
Eric Mendelsohn
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 21 Aug 18
Posts: 1
Credit: 0
RAC: 0
Message 67718 - Posted: 21 Aug 2018, 17:42:13 UTC

Hello all,

My name is Eric Mendelsohn, and I am a new developer working on the code for MilkyWay@Home. I am currently working on testing the impact of several parameters implicit within the N-body integrator used in MilkyWay@Home. I am also working on adding new gravitational potential models for the simulated bulge, disk, and halo of the Milky Way Galaxy. If you have any questions involving N-body, I will do my best to answer them as accurately as possible.

Thank you for your support!
____________

Profile Bill F
Send message
Joined: 4 Jul 09
Posts: 12
Credit: 1,170,900
RAC: 524

Message 67719 - Posted: 22 Aug 2018, 2:38:52 UTC

Eric

Welcome aboard !!!

I am looking forward to your efforts to improve the MilkyWay Project. A lot of the volunteer crunchers out here have been with the project for years. We are committed to the project and have weathered a few storms and hick ups.

Please be as frequent as you can be with news and updates as they are always appreciated.

I understand that in code development you can't test every platform, but the more you can debug an application version before release the less pain we have out here.

Again welcome aboard

Bill F
Dallas TX
____________

Profile ab1vl-pi
Avatar
Send message
Joined: 14 Jun 18
Posts: 4
Credit: 208,855
RAC: 2,635

Message 67720 - Posted: 22 Aug 2018, 21:18:16 UTC

Have had bad luck with nbody WUs. My latest, which right now BOINC shows as using 2 cores, has actually started to *increment* (after running OK almost 3 hours) the Remaining time field. Previously, large nbody WUs were getting stuck and showing no processing at all. Modfits have been running normally with no errors on win7 and Lubuntu linux with linux boinc.

I'm going to abort it and stop downloading new MWAH nbody tasks. I reported previous nbody stalls to Sid but never heard anything back.

HTH

Dataman
Avatar
Send message
Joined: 5 Sep 08
Posts: 19
Credit: 141,180,098
RAC: 93,937

Message 67721 - Posted: 22 Aug 2018, 23:39:17 UTC

A hearty welcome Eric!!

I have not been been here for a bit due to frequent problems with the project. I saw there were new wu's so I moved 4 machines here to give them a bit of a go. So far I have had no problems with Win10 or Linux.

Cheers and good luck to you.
____________

Forester
Send message
Joined: 28 Jun 17
Posts: 4
Credit: 17,432,783
RAC: 65,489

Message 67725 - Posted: 25 Aug 2018, 19:36:03 UTC - in response to Message 67718.
Last modified: 25 Aug 2018, 19:47:12 UTC

Welcome. I have have posted this some time ago but I'll say it again. It would be much appreciated if this project wold have an yearly activity rapport (as a 10 page PDF (ex:1 best contributors and best teams 2 what we have simulated and what observations we have extracted 3 observations vs simulations 4 limitation of our simulations 5 sponsors and gratitude etc)) so that us the crunchers would have an idea of the results of our compute time. thank you. Much appreciate if you do not ignore this post.

Jim1348
Send message
Joined: 9 Jul 17
Posts: 18
Credit: 825,091
RAC: 4,117

Message 67726 - Posted: 26 Aug 2018, 6:04:42 UTC - in response to Message 67725.

A 10 page report? I think a few lines of how things are going four times a year is quite enough. There are usually no real scientific conclusions that can be drawn until the project is complete.

Forester
Send message
Joined: 28 Jun 17
Posts: 4
Credit: 17,432,783
RAC: 65,489

Message 67730 - Posted: 27 Aug 2018, 14:48:01 UTC - in response to Message 67726.

Anything is better then nothing. Well We are volunteers.. that doesn't mean that we do not have costs with maintenance and electricity... What we ask is a bit of respect in return. nothing more.

Jim1348
Send message
Joined: 9 Jul 17
Posts: 18
Credit: 825,091
RAC: 4,117

Message 67732 - Posted: 27 Aug 2018, 21:54:09 UTC - in response to Message 67730.

I agree. Too many projects ignore the crunchers entirely. We need to know that someone is alive at the other end, looking at the data. Beyond that, I don't need much until the results are ready. Then, a link to the publication or other results would be in order.

Mr McGill
Send message
Joined: 13 Nov 17
Posts: 4
Credit: 1,983,076
RAC: 4,811

Message 67733 - Posted: 27 Aug 2018, 22:04:20 UTC
Last modified: 27 Aug 2018, 22:04:45 UTC

Currently, periodic angle parameters θ and φ are not allowed to wrap from 3.14 to 0
or from 0 to 3.14 making it possible for optimizations to get stuck on a boundary of these
parameters even though the best value lies within the constraints. If the best value for one
of these parameters is on the opposite side of the periodic boundary from the current search
neighborhood, the optimizer will continue approaching the boundary causing it to get stuck.


You mean to tell me: that after all this time: all you had to do was allow the angle to roll out to 3.15 by performing a quick 0'ing of any run attempted above Pi? Quite possibly more complex to implement: but it would have solved so many N-Body Lockups wouldn't it?

ultraZ
Avatar
Send message
Joined: 15 Dec 09
Posts: 8
Credit: 318,703,303
RAC: 322,665

Message 67737 - Posted: 29 Aug 2018, 3:29:07 UTC
Last modified: 29 Aug 2018, 3:30:42 UTC

Not having any problems with it taking 14 threads and 4 ati wu running concurrently.
____________

Profile Jake Weiss
Volunteer moderator
Project developer
Project tester
Project scientist
Send message
Joined: 25 Feb 13
Posts: 501
Credit: 34,647,251
RAC: 338

Message 67741 - Posted: 29 Aug 2018, 17:02:08 UTC

Hey Everyone,

Recently, we published a 100 page thesis on my work and a 30 page paper. Sidd is also working on an Nbody paper and thesis that should be done by the end of the year.

Thank you all for your continued support.

Jake

[VENETO] boboviz
Send message
Joined: 10 Feb 09
Posts: 36
Credit: 6,331,303
RAC: 229

Message 67758 - Posted: 31 Aug 2018, 15:45:01 UTC - in response to Message 67741.

Well done!!


Post to thread

Message boards : News : New MilkyWay@Home Developer


Main page · Your account · Message boards


Copyright © 2018 AstroInformatics Group