Skip to content

Latest commit

 

History

History
58 lines (33 loc) · 2.64 KB

README.markdown

File metadata and controls

58 lines (33 loc) · 2.64 KB
Raw

#cupSODA release 1.1.0

ABOUT

cupSODA is a black-box deterministic simulator of biological systems that exploits the remarkable memory bandwidth and computational capability of GPUs. cupSODA allows to efficiently execute in parallel large numbers of simulations, which are usually required to investigate the emergent dynamics of a given biological system under different conditions. cupSODA works by automatically deriving the system of ordinary differential equations from a reaction-based mechanistic model, defined according to the mass-action kinetics, and then exploiting the numerical integration algorithm, LSODA.

HOW TO CITE cupSODA

Nobile M.S., Cazzaniga P., Besozzi D., Mauri G.: GPU-accelerated simulations of mass-action kinetics models with cupSODA, Journal of Supercomputing, vol. 69, issue 1, pp.17–24, 2014

DEPENDENCIES

Just the Nvidia CUDA library (version >7.0).

COMPILATION

A cupSODA binary can be compiled on any supported architecture (e.g., GNU/Linux, Microsoft Windows, Apple OS/X) using the following compilation command:

nvcc kernel.cu cupSODA.cu input_reader.cpp stoc2det.cpp -gencode=arch=compute_20,code=compute_20 -O3 -o cupSODA --use_fast_math

The command above would create a binary executable file runnable on GPUs with at least a compute capability equal to 2.0. Please note that a specific compute capability, supporting additional functionality, can be targeted by using the gencode argument. For instance, to target the compute capability 3.5, the following argument can be passed to nvcc:

-gencode=arch=compute_35,code=compute_35

LAUNCHING CUPSODA

cupSODA is designed to be launched from the command line. The arguments are:

cupSODA input_folder blocks output_folder prefix gpu fitness memory_configuration debug

where

  • input_folder is the path to the directory containing the input model;
  • blocks is the number of CUDA blocks used to distribute the requested parallel threads;
  • output_folder is the path to the directory that will store the output dynamics of the simulations;
  • prefix is the file name of the output files. A number, corresponding the thread, will be automatically appended to the filename by cupSODA;
  • debug enables debug information: 1 outputs everything, 2 outputs everything but LSODA istate values for each thread.

Further information about the gpu, fitness and memory_configuration arguments, along with the specifications of the input files, can be found at the following address:

https://docs.google.com/document/d/1gPq-mYk-IP-bVmiMZewGPmTJ6nMCH8al1nNr7OaBsv4/edit?usp=sharing

LICENSE

BSD License

CONTACT

[email protected]