PDA

View Full Version : MilkyWay@Home



Bok
11-11-2007, 10:34 AM
New BOINC project MilkyWay@Home (http://milkyway.cs.rpi.edu/milkyway/)

Not too many wu's available just yet, I've just grabbed a few to try it out whilst Cosmology is unavailable...

team is http://milkyway.cs.rpi.edu/milkyway/team_display.php?teamid=23


xml is exported, so stats are at http://stats.free-dc.org/new/projpage.php?proj=mil

They are also here (http://stats2.free-dc.org/project_teams?proj=mil) for anyone wanting to take a peek at the newer stats being written under zope....

Bok :cheers:

LAURENU2
11-11-2007, 02:09 PM
:umm:Why are we at 23rd place:Pokes: that just is not right:swear:

.....Go.......Angus.......Go
:allhail:

IronBits
11-11-2007, 02:28 PM
SCHWEET !!!
:rock:


They are also here (http://stats2.free-dc.org/project_teams?proj=mil) for anyone wanting to take a peek at the newer stats being written under zope....
Bok :cheers:

sTs950
11-11-2007, 02:55 PM
thanks bok. I'm going to jump on this while cosmo is down also.

Fozzie
11-11-2007, 04:31 PM
let's go looking for top ten and beyond. :stomp:

New stats look great Bok :notworthy:

8 wus per machine get crunched very quickly

Helix_Von_Smelix
11-11-2007, 05:16 PM
it looks like this will be a very long running project, so will set one box plugging away for ever and a day.

I will Finish up other work and swap over:cheers:

LAURENU2
11-11-2007, 07:59 PM
Now I busted this project too:bang:
:umm: They ran out of work

Paratima
11-11-2007, 08:09 PM
You must be a killer at parties!

"Hey, I know, let's play Scrabble!"

"Can't. Lauren ate all the vowels."

:jester:

LAURENU2
11-11-2007, 08:31 PM
You must be a killer at parties!

"Hey, I know, let's play Scrabble!"

"Can't. Lauren ate all the vowels."

:jester:
Sory it is a old habit from way back in my Web Testing days:rock:

LAURENU2
11-12-2007, 01:04 AM
let's go looking for top ten and beyond. :stomp:

New stats look great Bok :notworthy:

8 wus per machine get crunched very quickly
I think we just hit 10th on there website

:umm: We did 13 today can we do 10 tomarrow:confused:

IronBits
11-12-2007, 09:50 AM
The work units only last ~3 minutes each, so I'm sure it will be difficult for the server to keep up, kinda like Eon, good thing to have running along side something else Boinc related.

http://milkyway.cs.rpi.edu/milkyway/server_status.php

LAURENU2
11-12-2007, 10:15 AM
Bok your stat page for this project has the wrong link on it http://milkyway.cs.rpi.edu/milkyway/stats/
it should be http://milkyway.cs.rpi.edu/milkyway

Bok
11-12-2007, 10:23 AM
oops. That's what I get for cutting and pasting without looking :)

Fixed...

LAURENU2
11-12-2007, 03:12 PM
I think we just hit 10th on there website

:umm: We did 13 today can we do 10 tomorrow:confused:

:thumbs: We have jumped 8 places so far today :thumbs: How many more jumps can we do today

Bok
11-12-2007, 04:21 PM
Note that I'm getting quite a few wu's crash out, leaving BOINC spinning until I close the error window on each machine....

Better check those.. Boincview shows each wu up in yellow in these cases..

Bok :cheers:

*EDIT* seems they are aware of it and it might be fixed now...http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=52

Angus
11-12-2007, 05:19 PM
:bigtrain: :train:

I feel so.... run over.


Thanks for the team boost :rock:

LAURENU2
11-13-2007, 12:26 AM
Note that I'm getting quite a few wu's crash out, leaving BOINC spinning until I close the error window on each machine....

Better check those.. Boincview shows each wu up in yellow in these cases..

Bok :cheers:

*EDIT* seems they are aware of it and it might be fixed now...http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=52
Ya same here since late last night

Also the scoreing on this project is all messed up

This computer gets 1,291.38 points for doing only 159 jobs GenuineIntel
Intel(R) Xeon(R) CPU X5355 @ 2.66GHz [x86 Family 6 Model 15 Stepping 7] Microsoft Windows XP
Professional Edition, Service Pack 2, (05.01.2600.00)
And I see it is repeatably

We only get 1 to 3 points on norm NOT RIGHT:swear:

LAURENU2
11-13-2007, 08:45 PM
:Pokes: Bok what happened your Stats are not updating :Pokes: But there stats are working

Bok
11-13-2007, 10:20 PM
Their xml stats aren't though...

http://milkyway.cs.rpi.edu/milkyway/stats/

/me thinks they didn't set up the cronjob correctly...

LAURENU2
11-14-2007, 01:02 AM
Their xml stats aren't though...

http://milkyway.cs.rpi.edu/milkyway/stats/

/me thinks they didn't set up the cronjob correctly...
A cronjob :umm: Is that legal here in the U-States=:harhar:

LAURENU2
11-14-2007, 01:11 AM
:hiya:I Just Checked MilkyWay@Home stats and :woot: We Are #1:banana: :hifi: :banana:

Bok
11-14-2007, 12:22 PM
:rock:

IronBits
11-14-2007, 06:01 PM
:banana:

Bok
11-14-2007, 06:29 PM
some info just posted on the message boards by the project admin...



I'm going to be starting a few more searches now that'll hopefully run while i'm out of town until saturday. the results are looking great :)

Right now we're looking into two different things:

1. The effect of the asynchronous return time of work units on the quality of the genetic searches we're running. If you'd like to read more about the asynchronous genetic search we're running, we have some papers we sent out recently to workshops/conferences detailing the search -- but we only had results using the new BlueGene supercomputer here at RPI, and our local RPI Grid of clusters.

The papers can be found here:
http://www.cs.rpi.edu/~deselt/desell_genetic_ppam_2007.pdf
http://www.cs.rpi.edu/~deselt/hcw2008.pdf

What we're hoping to do with this information is determine what effect a really heterogeneous computing environment has on our asynchronous genetic search, and figure out ways to optimize the search for these types of environments.

2. The astronomy program we're running currently calculates an integral over the wedge of the sky we're modeling. Since it's not possible to actually integrate the function we're interested in, we have to calculate the integral the old fashioned (and expensive way) of calculating a bunch of points and calculating the area between them. The next few searches we plan on doing will calculate the integral with varying accuracy - so expect work units of varying times (probably 8 minutes to over an hour). With these results we'll be able to figure out what effect the integral accuracy has on the rate our genetic search converges to a solution, and what effect the integral accuracy has on the quality of the answer.


Some other things that we have in the works is that i've been working on an asynchronous particle swarm algorithm to use (as opposed to asynchronous genetic search). For a brief overview of what particle swarm optimization is, wikipedia isn't too shabby: http://en.wikipedia.org/wiki/Particle_swarm_optimization . We'll be able to contrast and compare these two searches, which should be interesting.

Also, we're using the results to compare doing this work on BOINC to the BlueGene supercomputer at RPI. Who knows, with enough participants in the project -- you guys might beat out the 7th fastest supercomputer in the world! :D (see: http://www.top500.org/lists/2007/06 )

LAURENU2
11-14-2007, 11:42 PM
you guys might beat out the 7th fastest supercomputer in the world! (see: http://www.top500.org/lists/2007/06 )

Well if anyone can do it Free-DC can:thumbs:

Helix_Von_Smelix
11-15-2007, 03:14 AM
Note that I'm getting quite a few wu's crash out, leaving BOINC spinning until I close the error window on each machine....

Better check those.. Boincview shows each wu up in yellow in these cases..

Bok :cheers:

*EDIT* seems they are aware of it and it might be fixed now...http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=52

To stop BOINC hanging with an error >> control panel >> system >> advanced tab >> (at the bottom) error reporting >> disable & don't notify

:cheers:

Bok
11-15-2007, 07:32 AM
cool. Never knew about that one!

Bok :cheers:

Paratima
11-15-2007, 10:03 AM
OK, I'm sucked into yet another priceless project! :rotfl:

This gets one core, haven't decided which yet. Currently running on one half of AMD 6000 X2, may switch to laptop.

BOK, you can save yourself some interruptions by, on that window, checking Don't Notify Microsoft. That is, unless you'd like your error reports piled on top of the billions of other unregarded, unwanted and uninspected error reports from the rest of the M$ world. This is the switch that brings up the neat little window, "Would you like to send a report of this error to Oregon?" or some such. Waste of time.