Welcome to MilkyWay@home

What am I actually doing?

Message boards : MilkyWay@home Science : What am I actually doing?
Message board moderation

To post messages, you must log in.

AuthorMessage
Robin95

Send message
Joined: 21 Jul 13
Posts: 2
Credit: 545,069
RAC: 0
Message 59800 - Posted: 4 Sep 2013, 8:08:09 UTC

Hello

Since MilkyWay@Home 1.02 OpenCL tasks are finished very fast, I really want to know what I actually contribute by crunching 1 task and sending it back to the project. What does 1 task do and what impact does it have on the whole project?
Is it like I replicate a small piece of the universe or what? Please help me out ;D

Greetings.
ID: 59800 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Matthew
Volunteer moderator
Project developer
Project scientist

Send message
Joined: 6 May 09
Posts: 217
Credit: 6,856,375
RAC: 0
Message 59871 - Posted: 10 Sep 2013, 19:59:34 UTC - in response to Message 59800.  

See Part III of the science page:

http://milkyway.cs.rpi.edu/milkyway/science.php#part3

Each work unit is a single "guess" at the parameters that best describe the part of the Galaxy being studied. The crunching that you are doing is a massive computation that compares a single set of parameters to a single set of data; the math that we do is really, really complicated (see Nathan Cole's thesis), and has to be done for each star in the data, so it requires a lot of crunching power.

Each "guess" is then compared to the data then returned to the server. The server then uses advanced statistics to determine which "guesses" would be the best to try next. Eventually, the algorithm converges to a set of parameters that the absolute ("global") best, or the "maximum likelihood" that those parameters describe the data. These bests then have to be checked by a human to see if they are reasonable, as sometimes the algorithm finds a solution that gives a good likelihood, but makes no physical sense (such as finding a tidal stream the size of the MilkyWay galaxy...). Since we use only a portion of our data at a time, sometimes local effects can skew the algorithm - it's not perfect! But we are constantly making it better - when a new version of the app(s) is released, it usually includes a new feature designed to improve the search.
ID: 59871 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Robin95

Send message
Joined: 21 Jul 13
Posts: 2
Credit: 545,069
RAC: 0
Message 59884 - Posted: 12 Sep 2013, 10:10:47 UTC

Thank you!
ID: 59884 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Mary Mags

Send message
Joined: 7 Nov 13
Posts: 4
Credit: 141,909
RAC: 0
Message 60364 - Posted: 11 Nov 2013, 20:18:23 UTC

Yes, thank you, Matthew.

- However, it still sounds like you have tons of work in your in-basket. (if you're still doing the math to check on the computer).

- Does each new volunteer take a sheet of paper OUT of that basket - or do we just add to the pile?

- It would be nice to think we're (I'm) helping in some way. I believe I am but I think you need an assistant, or 2... or 2 million.

- I think it's fascinating and I thank you and your team for your efforts. Wish I could do more.
ID: 60364 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Matthew
Volunteer moderator
Project developer
Project scientist

Send message
Joined: 6 May 09
Posts: 217
Credit: 6,856,375
RAC: 0
Message 60437 - Posted: 20 Nov 2013, 0:11:15 UTC - in response to Message 60364.  

- Does each new volunteer take a sheet of paper OUT of that basket - or do we just add to the pile?


Every user that signs up is helping with work that needs to get done - so every user makes things better!

... but I think you need an assistant, or 2... or 2 million.


Haha, yes, I could use a few dozen assistants :). Jeff Thompson and Jake Weiss have been doing a great job, though, and Jake Bauer has been improving the N-body test application every week.

The real draw on my time right now is teaching classes, searching for jobs (I received my PhD last Summer), wrapping up some in-house projects, behind-the-scenes administrative work, keeping MW running, public outreach stuff, getting the research I need done finished on MW, creating badges for MW, and eventually finishing the computer science part of the science page. Oh, and trying to stay sane ;).

- I think it's fascinating and I thank you and your team for your efforts. Wish I could do more.


Any help you can provide is great! We hope to have our donor page up and running again sometime soon, and I would like to start a "Cafepress"-style store for MilkyWay@home t-shirts and stuff. We're working hard to make this project even cooler than it is. :)
ID: 60437 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
dan

Send message
Joined: 7 Dec 13
Posts: 3
Credit: 373,601
RAC: 0
Message 60597 - Posted: 15 Dec 2013, 15:08:24 UTC

So when you found a new job, what will happen to this project?
Congratulations on the phd!
ID: 60597 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[TA]Assimilator1
Avatar

Send message
Joined: 22 Jan 11
Posts: 375
Credit: 64,657,871
RAC: 0
Message 60719 - Posted: 11 Jan 2014, 17:28:02 UTC - in response to Message 60597.  

Good question!
Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7
ID: 60719 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : MilkyWay@home Science : What am I actually doing?

©2024 Astroinformatics Group