Welcome to MilkyWay@home

Nbody Science


Advanced search

Message boards : MilkyWay@home Science : Nbody Science
Message board moderation

To post messages, you must log in.

AuthorMessage
Colin Rice
Volunteer moderator
Project developer
Project tester
Project scientist
Avatar

Send message
Joined: 5 Oct 11
Posts: 3
Credit: 485
RAC: 0
1 credit badge9 year member badge
Message 51309 - Posted: 6 Oct 2011, 0:09:03 UTC

Hi,

I'm Colin Rice an undergraduate at RPI who has been working with mw@home for about a semester now. I'm posting to give a little insight into the nbody simulations which I recently took over.

A N-Body simulation is a simulation where you are using multiple small bodies and applying forces between them to simulate some sort of physical system. In our case we are simulating a dwarf galaxy orbiting the milkyway.

We are currently investigating the makeup of dark matter within dwarf galaxies. Due to inconsistent angular velocities we think there may be dark matter inside of dwarf galaxies. However it is not conclusively proven.

Our goal is to determine if dark matter inside of a dwarf galaxy causes a distinctive orbit when the galaxy is then pulled into an orbit about the milkyway. We are doing this in two steps.

1)We are currently running simulations to see if the N-Body system can recover the initial parameters of a dwarf galaxy which we generated with dark and light matter, and then evolved in time.

Specifically, when an nbody runs the server has generated a trial set of parameters and is evolving them to see if they match the observed histogram. It then evaluates how well they worked and creates new runs based on the current runs fitness. It only evaluates the fitness based on the light matter. However, it used both light and dark matter when running the simulation. This is done because we can only see the light matter in real world data and that is what this system is designed to process.

2) The next step after we confirm that the technique works via test runs is to put real world data into the simulation and see if we can figure out the initial parameters for the dark matter inside the dwarf galaxies. We can't just plug in the initial data because we are unsure as to whether the system works.
ID: 51309 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileBill F
Avatar

Send message
Joined: 4 Jul 09
Posts: 17
Credit: 5,571,505
RAC: 5,088
5 million credit badge10 year member badge
Message 67898 - Posted: 17 Nov 2018, 9:55:45 UTC - in response to Message 51309.  

Hi,

I'm Colin Rice an undergraduate at RPI who has been working with mw@home for about a semester now. I'm posting to give a little insight into the nbody simulations which I recently took over.

A N-Body simulation is a simulation where you are using multiple small bodies and applying forces between them to simulate some sort of physical system. In our case we are simulating a dwarf galaxy orbiting the milkyway.

We are currently investigating the makeup of dark matter within dwarf galaxies. Due to inconsistent angular velocities we think there may be dark matter inside of dwarf galaxies. However it is not conclusively proven.

Our goal is to determine if dark matter inside of a dwarf galaxy causes a distinctive orbit when the galaxy is then pulled into an orbit about the milkyway. We are doing this in two steps.

1)We are currently running simulations to see if the N-Body system can recover the initial parameters of a dwarf galaxy which we generated with dark and light matter, and then evolved in time.

Specifically, when an nbody runs the server has generated a trial set of parameters and is evolving them to see if they match the observed histogram. It then evaluates how well they worked and creates new runs based on the current runs fitness. It only evaluates the fitness based on the light matter. However, it used both light and dark matter when running the simulation. This is done because we can only see the light matter in real world data and that is what this system is designed to process.

2) The next step after we confirm that the technique works via test runs is to put real world data into the simulation and see if we can figure out the initial parameters for the dark matter inside the dwarf galaxies. We can't just plug in the initial data because we are unsure as to whether the system works.


Can we get an update on this thread ? Does the system work and are we feeding any real data yet using it ?

Thanks
Bill F
ID: 67898 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
greg_be

Send message
Joined: 18 Aug 09
Posts: 100
Credit: 12,151,374
RAC: 21,463
10 million credit badge10 year member badge
Message 69062 - Posted: 19 Sep 2019, 5:42:18 UTC

Gee guys...no new info in over 8 years and no reply to a person in almost a year.
Love the communication.
ID: 69062 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
AquaBoy

Send message
Joined: 30 Apr 18
Posts: 3
Credit: 23,006,977
RAC: 4,937
20 million credit badge2 year member badge
Message 69087 - Posted: 20 Sep 2019, 10:29:17 UTC
Last modified: 20 Sep 2019, 10:30:50 UTC

Unfortunately, the project founders don't seem to have any respect for our enthusiastic and altruistic crunching community, since they say nothing in response to our questions. Yet, why only one article has been published by the end of this year? It's very little as compared to the previous years. Maybe, you experience some difficulties or have a lot of work to do ahead?
ID: 69087 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
AquaBoy

Send message
Joined: 30 Apr 18
Posts: 3
Credit: 23,006,977
RAC: 4,937
20 million credit badge2 year member badge
Message 69101 - Posted: 21 Sep 2019, 8:04:48 UTC

In any way, we hope to hear some positive news from you soon. Good luck in your not easy work!
ID: 69101 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileJonathan Melusky
Avatar

Send message
Joined: 20 May 18
Posts: 5
Credit: 643,958
RAC: 883
500 thousand credit badge2 year member badge
Message 69872 - Posted: 29 May 2020, 19:45:25 UTC

The last 4 years have been brutal to science budgets in the US. In 2018 alone, the budget for NASA was cut by over $19 billion so money and grants for Milkyway public outreach most likely took a hit in funding. In 2017, the NASA budget was also cut, but not by as much. Here is a quote from an article I found.

"Out of all the federal agencies that received budget plans from President Donald Trump yesterday, NASA fared pretty well. The space agency is only facing a 0.8 percent cut in its overall budget — a relatively mild change compared to the Environmental Protection Agency, which could see its funding slashed by 31 percent. But packed within NASA’s small budget decrease are some pretty sizable cuts. A few major upcoming missions are canceled, and NASA’s entire education program, which is responsible for outreach and grants, is eliminated. The budget request also proposes wasting technologies already in space."
https://boincstats.com/signature/-1/user/4394448/sig.png
ID: 69872 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileEric Mendelsohn
Volunteer moderator
Project administrator
Project developer
Project scientist

Send message
Joined: 21 Aug 18
Posts: 51
Credit: 5,044,810
RAC: 317
5 million credit badge2 year member badge
Message 69897 - Posted: 8 Jun 2020, 19:36:57 UTC

Hey everyone,

It's been a while since we've had an update to this thread. A lot has been happening with MilkyWay@home, and I'd like to discuss some of the new things that we are doing on here.
Currently running on MilkyWay@home, we are optimizing the Orphan Stream's Dwarf Galaxy progenitor using data from the Sloan Digital Sky Survey (SDSS) as well as data from the Dark Energy Camera to fit some of the southern tail of the Orphan Stream. In the past, we were testing MW@home with simulated data to see if we could accurately pull the correct mass and radial profile of a sample dwarf galaxy from its tidal streams. After fixing several bugs with the Earth Mover Distance code and collecting data in areas where we had none, we are finally ready to use our code on the real thing.

In addition to the new runs, we are getting ever closer to releasing another version of MW@home with a number of new features. One of the biggest features we are adding is the capacity for MW@home to fit the orbits of the tidal streams we optimize. In past versions, the orbit (defined by two positional parameters and a 3-dimensional velocity parameters) had to be manually set in the lua file sent with all of the workunits. Using stellar data, we would calculate the five parameters needed to have the orbit of our dwarf galaxy trace the path of the observed tidal streams. This, however, is not always a reliable method for selecting the orbit. Dwarf galaxies are not infinitesimal points, and thus trace out several self-interacting orbits which phase mix as tidal forces cause segments of the dwarf galaxy to accelerate forward or lag behind. Also, in some cases, the orientation of a tidal stream may not reflect the true orbital path it takes. In the Orphan Stream's case, the velocity of stars in its southern tail move PERPENDICULARLY to the tidal stream due to gravitational tugs from the Large Magellanic Cloud. To account for this variability and chaos, in addition to fitting the mass and radial profile of the progenitor dwarf galaxy, we also plan on simultaneously fitting its orbit.

Another new feature in the upcoming version is the addition of the LMC. New estimates of the LMC's mass place it at around 10% of the Milky Way's total mass. This contributes a significant non-static component to the Milky Way system that needs to be accounted for in simulation. However, adding an LMC of this mass is not as simple as inserting a point particle with the appropriate mass into our simulation. As the LMC is pulled towards the Milky Way Galaxy, our Galaxy is similarly pulled by the LMC, shifting our system into an accelerating frame of reference. As such, we are required to correct the positions and velocities of all particles for each time step in order to account for shifts in the Milky Way's reference frame.

The last feature we are still working on adding in MW@home is GPU processing for nbody. Just as Separation in MW@home uses GPU applications to quickly pull streams from the sky, we plan on implementing GPU computations to speed up our lengthy simulations. It's no secret that NBody applications have been taking longer to run between the increase in bodies and the addition of more complicated potentials. Plus, the orbit fitting and LMC calculations only add to the runtime. While we are still fixing some bugs, we are hopeful that adding GPU compatibility will help everyone save time and speed through these calculations.

All in all, we've been busy at RPI pulling new stellar data for optimizations and debugging the new upcoming features. We apologize for not being as active in these threads as we were in the past, and we will attempt to keep the updates coming as new breakthroughs are made. Thank you all for your patience and support.

-Eric
ID: 69897 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Trotador

Send message
Joined: 5 Oct 13
Posts: 5
Credit: 615,804,730
RAC: 593,457
500 million credit badge7 year member badge
Message 69900 - Posted: 9 Jun 2020, 6:22:19 UTC - in response to Message 69897.  

Great!, thanks for the update!
ID: 69900 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileKeith Myers
Avatar

Send message
Joined: 24 Jan 11
Posts: 335
Credit: 215,103,902
RAC: 320,641
200 million credit badge9 year member badgeextraordinary contributions badge
Message 69901 - Posted: 9 Jun 2020, 6:53:20 UTC

Thanks for the update Eric. Much appreciated to be brought into the loop again. Looking forward to the new gpu appplications.
ID: 69901 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Falconet

Send message
Joined: 9 Mar 09
Posts: 3
Credit: 36,840
RAC: 0
10 thousand credit badge10 year member badge
Message 69922 - Posted: 14 Jun 2020, 13:12:03 UTC

Thank you for the update.
ID: 69922 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileZa69uzZ
Avatar

Send message
Joined: 6 Apr 12
Posts: 42
Credit: 3,215,609
RAC: 0
3 million credit badge8 year member badge
Message 69928 - Posted: 17 Jun 2020, 19:01:49 UTC - in response to Message 69872.  

The last 4 years have been brutal to science budgets in the US. In 2018 alone, the budget for NASA was cut by over $19 billion so money and grants for Milkyway public outreach most likely took a hit in funding. In 2017, the NASA budget was also cut, but not by as much. Here is a quote from an article I found.

"Out of all the federal agencies that received budget plans from President Donald Trump yesterday, NASA fared pretty well. The space agency is only facing a 0.8 percent cut in its overall budget — a relatively mild change compared to the Environmental Protection Agency, which could see its funding slashed by 31 percent. But packed within NASA’s small budget decrease are some pretty sizable cuts. A few major upcoming missions are canceled, and NASA’s entire education program, which is responsible for outreach and grants, is eliminated. The budget request also proposes wasting technologies already in space."



BOINC has a crypto crunching fundraiser to help scientific projects
ID: 69928 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileChooka
Avatar

Send message
Joined: 13 Dec 12
Posts: 73
Credit: 1,089,356,640
RAC: 4,579,997
1 billion credit badge7 year member badgeextraordinary contributions badge
Message 70005 - Posted: 28 Jul 2020, 8:21:25 UTC

Thanks for the update Eric.
I must admit, I was beginning to wonder why I continue to crunch M@H vs Einstein@Home.

ID: 70005 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : MilkyWay@home Science : Nbody Science

©2020 Astroinformatics Group