Detail publikace

Towards Accelerated Computation of Atmospheric Equations using CUDA

Originální název

Towards Accelerated Computation of Atmospheric Equations using CUDA

Anglický název

Towards Accelerated Computation of Atmospheric Equations using CUDA

Jazyk

en

Originální abstrakt

Main objective of this paper is to outline possible ways how to achieve a substantial acceleration in case of advection-diffusion equation (A-DE) calculation, which is commonly used for a description of the pollutant behavior in atmosphere. A-DE is a kind of partial differential equation (PDE) and in general case it is usually solved by numerical integration due to its high complexity. These types of calculations are time consuming thus the main idea of our work is to adopt CUDA platform and commodity GPU card to do the calculations in a faster way. The solution is based on method of lines with 4th order Runge-Kutta scheme to handle the integration. As a matter of fact, the selected approach involves number of auxiliary variables and thus the memory management is critical in order to achieve desired performance. We have implemented several possible solutions that use different memory access schemes. Detailed evaluation is provided in this paper where the obtained results show a tremendous processing speed up in comparison to CPU.

Anglický abstrakt

Main objective of this paper is to outline possible ways how to achieve a substantial acceleration in case of advection-diffusion equation (A-DE) calculation, which is commonly used for a description of the pollutant behavior in atmosphere. A-DE is a kind of partial differential equation (PDE) and in general case it is usually solved by numerical integration due to its high complexity. These types of calculations are time consuming thus the main idea of our work is to adopt CUDA platform and commodity GPU card to do the calculations in a faster way. The solution is based on method of lines with 4th order Runge-Kutta scheme to handle the integration. As a matter of fact, the selected approach involves number of auxiliary variables and thus the memory management is critical in order to achieve desired performance. We have implemented several possible solutions that use different memory access schemes. Detailed evaluation is provided in this paper where the obtained results show a tremendous processing speed up in comparison to CPU.

BibTex


@inproceedings{BUT30199,
  author="Václav {Šimek} and Radim {Dvořák} and František {Zbořil} and Jiří {Kunovský}",
  title="Towards Accelerated Computation of Atmospheric Equations using CUDA",
  annote="Main objective of this paper is to outline possible ways how to achieve
a substantial acceleration in case of advection-diffusion equation (A-DE)
calculation, which is commonly used for a description of the pollutant behavior
in atmosphere. A-DE is a kind of partial differential equation (PDE) and in
general case it is usually solved by numerical integration due to its high
complexity. These types of calculations are time consuming thus the main idea of
our work is to adopt CUDA platform and commodity GPU card to do the calculations
in a faster way. The solution is based on method of lines with 4th order
Runge-Kutta scheme to handle the integration. As a matter of fact, the selected
approach involves number of auxiliary variables and thus the memory management is
critical in order to achieve desired performance. We have implemented several
possible solutions that use different memory access schemes. Detailed evaluation
is provided in this paper where the obtained results show a tremendous processing
speed up in comparison to CPU.",
  address="IEEE Computer Society",
  booktitle="Proceedings of Eleventh International Conference on Computer Modelling and Simulation",
  chapter="30199",
  doi="10.1109/UKSIM.2009.25",
  edition="NEUVEDEN",
  howpublished="print",
  institution="IEEE Computer Society",
  year="2009",
  month="march",
  pages="449--454",
  publisher="IEEE Computer Society",
  type="conference paper"
}