Commit d2573609 authored by Volker Springel's avatar Volker Springel
Browse files

added license file

parent f69d0401
This list shows people who have repository access to GADGET-4, and
contribute actively to its development.
Name Email Institution
Volker Springel MPA
Ruediger Pakmor MPA
Oliver Zier MPA
Martin Reinecke MPA
This diff is collapsed.
......@@ -3,54 +3,32 @@
___ __ ____ ___ ____ ____ __
/ __) /__\ ( _ \ / __)( ___)(_ _)___ /. |
( (_-. /(__)\ )(_) )( (_-. )__) )( (___)(_ _)
\___/(__)(__)(____/ \___/(____) (__) (_)
GADGET-4 is a massively parallel code for N-body/hydrodynamical
cosmological simulations. It is a flexible code that can be applied to
a variety of different types of simulations, offering a number of
sophisticated simulation algorithms. An account of the numerical
algorithms employed by the code is given in the original code paper,
subsequent publications, and this documentation.
GADGET-4 was written mainly by
[Volker Springel](, with
important contributions and suggestions being made by numerous people,
including [Ruediger Pakmor](,
[Oliver Zier](, and
[Martin Reinecke](
To build a local copy of the documentation:
- run `doxygen`
- open `doxygen/html/index.html` in a browser
See the file DEVELOPERS.
* Maximum number of particles per MPI-rank is restricted to 2^31 ~ 2 Billion
* Maximum number of groups/subhalos per MPI-rank is restricted to 2^31 ~ 2 Billion
* The total number of particles, and the total number of groups/subhalos, can however
be much larger than this if a sufficiently large number of MPI ranks is used.
* Maximum particle number in a single group or subhalo is for the default setting 2^31 ~ 2 billion,
but this can be enlarged if needed by setting the option FOF_ALLOW_HUGE_GROUPLENGTH
* The maximum number of subhalos per group is restricted nevertheless to 2^31 ~ 2 billion.
* If the legacy output formats 1/2 are used, each block in the output is restricted to 2 Gbyte in
size or less. This also means that the maximum number of particles per single output file, and the
maximum number of groups/subhalos per single group catalogue file can be at most of order 2 Billion,
normally substantially less than that. While this restriction can be circumvented by splitting output
over enough files, it provides one further motivation to use the HDF5 formay!
* Maximum number of lightcones: 256
* If more MPI-ranks than PMGRID are used, not all MPI tasks get anything to do in FFT calculations, which severely
compromises achievable work-load and memory balance. In this case, the column-based FFT algorithm is
a way out (it incurs higher cost however, as twice as many transpose operations are needed)
For documentation of the code as well as the code paper, please refer
to the [code's web-site](
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment