Abstract
An implementation with the CUDA technology in a single and in several graphics processing units (GPUs) is presented for the calculation of the forward modeling of gravitational fields from a tridimensional volumetric ensemble composed by unitary prisms of constant density. We compared the performance results obtained with the GPUs against a previous version coded in OpenMP with MPI, and we analyzed the results on both platforms. Today, the use of GPUs represents a breakthrough in parallel computing, which has led to the development of several applications with various applications. Nevertheless, in some applications the decomposition of the tasks is not trivial, as can be appreciated in this paper. Unlike a trivial decomposition of the domain, we proposed to decompose the problem by sets of prisms and use different memory spaces per processing CUDA core, avoiding the performance decay as a result of the constant calls to kernels functions which would be needed in a parallelization by observations points. The design and implementation created are the main contributions of this work, because the parallelization scheme implemented is not trivial. The performance results obtained are comparable to those of a small processing cluster.
Highlights
In recent years the number of publications about parallel computing applications using the graphics processing units (GPUs) architecture has remarkably increased
Unlike a trivial decomposition of the domain, we proposed to decompose the problem by sets of prisms and use different memory spaces per processing Compute unified device architecture (CUDA) core, avoiding the performance decay as a result of the constant calls to kernels functions which would be needed in a parallelization by observations points
Since the architecture of the GPU is different to that of a conventional CPU, the programming paradigm should be changed. This had led to the development of a new research field within scientific computing which explores the performance of the GPU to general purpose applications, such as acoustic simulation [3], propagation of seismic waves [4], seismic migration [5], molecular engineering [6], fluid dynamics [7], even for astrophysical simulations [8] and many other implementations
Summary
In recent years the number of publications about parallel computing applications using the GPUs architecture has remarkably increased. When these difficulties are present in the application design, it is necessary to modify the strategy to achieve a good performance Implementing this application in CUDA for the calculation of the direct model through a prisms ensemble represents a big challenge; this is because the memory region where the calculations are made (observation grid) is shared and requires different memory allocations to avoid the data coherency problems produced when two or more processing cores access the same memory location at the same time. This paper is organized as follows: in Section 2 we present the characteristics of the CUDA platform and the tools we used, in Section 3 the application design is explained, in Section 4 we present some numerical experiments that were made, in Section 5 we detail the validation of the code for double as well as single precision, and in Section 6 we compare against a 29-nodes cluster and we present our conclusions
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
