Numerical problems to be solved in astronomy

Hello everyone,

I wanted to experiment with parallel programming for pcs. And I was
wondering if there was a problem dealing with astronomy that I can try
solving by using parallel programming. I am looking for problems that
deal with equations of any type basically like calculating a planets
position for the next 100 years or so. But I would like to see if
there are other things other can calculating an ephermeris that I can
do that would be interesting and on a topic that everybody would like
some computed data on. Any pointers to websites or books would be very
helpful.

Thanking you all,

Zefram Cochrane

In article ,
(Zefram Cochrane) wrote:

Hello everyone,

I wanted to experiment with parallel programming for pcs. And I was
wondering if there was a problem dealing with astronomy that I can try
solving by using parallel programming. I am looking for problems that
deal with equations of any type basically like calculating a planets
position for the next 100 years or so. But I would like to see if
there are other things other can calculating an ephermeris that I can
do that would be interesting and on a topic that everybody would like
some computed data on. Any pointers to websites or books would be very
helpful.

Thanking you all,

Zefram Cochrane

Computers are so fast today that we can calculate simple things like an
ephemeris without taking up a lot of time. To really get into parallel
programming and cluster work, you're looking at a lot more complicated
things, like the calculation of stellar structure models. You can go
look up some work by Travis Metcalfe and related WET articles for
information on that.

If you're just playing around, then you can pick just about anything and
go to it.

Reed

--
Dr. Reed L. Riddle
Associate Director of Whole Earth Telescope Operations
Iowa State University Department of Physics & Astronomy
Email: drriddle "at" qwest.net
Homepage: http://wet.physics.iastate.edu/~riddle/

"This life has been a test. If this had been an actual life, you would have
received instructions on where to go and what to do."
Angela Chase, "My so-called life"

Remove "DAMN SPAM" from my email address to reply.....

In article ,
Reed Riddle wrote:
To really get into parallel
programming and cluster work, you're looking at a lot more complicated
things, like the calculation of stellar structure models. You can go
look up some work by Travis Metcalfe and related WET articles for
information on that.

In fact there's a huge area of numerical simulations of astrophysical
phenomena that can eat just about any amount of CPU time you can throw
at it -- but to improve on existing codes you'd need a strong
background in computational physics. Go here
http://adsabs.harvard.edu/abstract_service.html
and type in `numerical modeling' in the title box and you'll get an
idea of what's out there.

One advantage of astronomical systems is that the physics is often
quite easily parallelizable. For example, I have some cluster-based
code that calculates inverse-Compton emissivity -- it's essentially
just a matter of dividing the object of interest into small chunks and
then handing the calculations out in turn to all the CPUs available.

Martin

[newsgroups trimmed]

--
Martin Hardcastle Department of Physics, University of Bristol
A little learning is a dangerous thing; / Drink deep, or taste not the
Pierian spring; / There shallow draughts intoxicate the brain ...

In sci.astro.research Martin Hardcastle wrote:
In fact there's a huge area of numerical simulations of astrophysical
phenomena that can eat just about any amount of CPU time you can throw
at it -- but to improve on existing codes you'd need a strong
background in computational physics. Go here
http://adsabs.harvard.edu/abstract_service.html
and type in `numerical modeling' in the title box and you'll get an
idea of what's out there.

One advantage of astronomical systems is that the physics is often
quite easily parallelizable. For example, I have some cluster-based
code that calculates inverse-Compton emissivity -- it's essentially
just a matter of dividing the object of interest into small chunks and
then handing the calculations out in turn to all the CPUs available.

Because different clusters use different types of software for
interprocessor communication, it's a bit messy -- and takes more
computer-science knowledge than a lot of (astro)physicists have --
to program such a parallel code in a portable and portably-efficient
way.

I'd like to point anyone thinking of trying this to an alternative,
namely the use of a "parallel toolkit environment" like
PETSc (http://www-fp.mcs.anl.gov/petsc/) or
Cactus (http://www.cactuscode.org).
These provide already-written subroutine libraries for common tasks
like time sequencing, parameter-file handling, bulk data I/O, solving
linear equations, etc. They generally provide almost-transparent
parallelism, usually implemented on top of MPI.

Cactus (which I use a lot in my work) is focused on the solution
of 3-D PDEs. It uses domain decomposition for parallelism, and does
a nice job of hiding the details of this from applications code.
To a reasonably approximation, it enables physicists to stay focused
on writing physics code, and not have to worry about the details of
parallelization.

As an example of the sort of astrophysics calculations people are
doing with Cactus, I'll mention a (very) recent paper by a group of
my colleagues (including my office-mate!), gr-qc/0403029:

Authors: L. Baiotti, I. Hawke, P.J. Montero, F. Loeffler, L. Rezzolla,
N. Stergioulas, J.A. Font, E. Seidel
Title: Three-dimensional relativistic simulations of
rotating neutron star collapse to a Kerr black hole
Abstract:
We present a new three-dimensional fully general relativistic
hydrodynamics code using high-resolution shock-capturing techniques and
a conformal-traceless formulation of the Einstein equations. Besides
presenting a thorough set of tests which the code passes with
very high accuracy, we discuss its application to the study of the
gravitational collapse of uniformly rotating neutron stars to Kerr
black holes. The initial stellar models are modelled as relativistic
polytropes which are either secularly or dynamically unstable and with
angular velocities which range from slow rotation to the mass-shedding
limit. We investigate the gravitational collapse by carefully studying
not only the dynamics of the matter, but also that of the trapped
surfaces, i.e. of both the apparent and event horizons formed during
the collapse. The use of these surfaces, together with the dynamical
horizon framework, allows for a precise measurement of the black hole
mass and spin. The ability to successfully perform these simulations for
sufficiently long times relies on excising a region of the computational
domain which includes the singularity and is within the apparent
horizon. The dynamics of the collapsing matter is strongly influenced
by the initial amount of angular momentum in the progenitor star and,
for initial models with sufficiently high angular velocities, the
collapse can lead to the formation of an unstable disc in differential
rotation. All of the simulations performed with uniformly rotating
initial data and a polytropic equation of state show no evidence of
shocks or the presence of matter on stable orbits outside the black
hole. Both conclusions may change if non-isentropic equations of state
are used and if the initial stellar models rotate differentially.

ciao,

--
-- Jonathan Thornburg
Max-Planck-Institut fuer Gravitationsphysik (Albert-Einstein-Institut),
Golm, Germany, "Old Europe" http://www.aei.mpg.de/~jthorn/home.html
"Washing one's hands of the conflict between the powerful and the
powerless means to side with the powerful, not to be neutral."
-- quote by Freire / poster by Oxfam

Zefram Cochrane wrote:
Hello everyone,

I wanted to experiment with parallel programming for pcs. And I was
wondering if there was a problem dealing with astronomy that I can try
solving by using parallel programming. I am looking for problems that
deal with equations of any type basically like calculating a planets
position for the next 100 years or so. But I would like to see if
there are other things other can calculating an ephermeris that I can
do that would be interesting and on a topic that everybody would like
some computed data on. Any pointers to websites or books would be very
helpful.

Thanking you all,

Zefram Cochrane

A highly regarded book is 'Astronomical Algorithms' by Jean Meeus -
although parallel processing will be overkill.

http://www.amazon.com/exec/obidos/tg...glance&s=books


--
25° 45' S
28° 12' E
GMT+2

Join the Planetary Society
http://www.planetary.org