A new CUDA-based GPU implementation of the two-dimensional Athena code

Wasiljew, A.; Murawski, K.

doi:10.2478/bpasts-2013-0023

Cited by 2 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The GPU accelerated algorithms constitute dynamically developing branch of computational techniques used extensively in many practical engineering problems (see e.g. [19][20][21][22]).…”

Section: Fast Parametric Transformsmentioning

confidence: 99%

Bulletin of the Polish Academy of Sciences: Technical Sciences

Lipiński¹,

Puchala²

2019

View full text Add to dashboard Cite

This paper proposes a novel method for digital image watermarking, in which watermarks are embedded in the domain of fast parametric transforms based on known spread spectrum approaches. Fast parametric transforms have the ability to adapt the forms of base vectors, which enables automatic selection of the domain of watermarking in relation to the pair: a marked image -a watermarking attack. The process of adapting the forms of fast parametric transforms is carried out with aid of the classical genetic algorithm with the fitting function based on the known measure of separability of watermarks. The effectiveness of the proposed method has been verified experimentally on the basis of the images of two classes, i.e. natural images and technical diagrams. The results taking into account both the efficiency of watermark embedding and the generated distortions in the marked images are summarized in tables and accompanied by an appropriate commentary.

show abstract

“…The GPU accelerated algorithms constitute dynamically developing branch of computational techniques used extensively in many practical engineering problems (see e.g. [19][20][21][22]).…”

Section: Fast Parametric Transformsmentioning

confidence: 99%

Bulletin of the Polish Academy of Sciences: Technical Sciences

Lipiński¹,

Puchala²

2019

View full text Add to dashboard Cite

show abstract

“…But a very limited of implementation details were reported in Lin's work. Recently, Wasiljew and Murawski [42] presented a GPU implementation of the 2D Athena code [35] and tested it on a NVIDIA GeForce GTX 460 graphics card. Our recent work [44] implemented a global MHD model on multiple GPUs on a single node, achieving a peak performance of 97.36 GFLOPS in double precision.…”

Section: Introductionmentioning

confidence: 99%

Efficient magnetohydrodynamic simulations on distributed multi-GPU systems using a novel GPU Direct–MPI hybrid approach

Wong

Aoki

Wong

2014

Computer Physics Communications

View full text Add to dashboard Cite

Modern graphics processing units (GPUs) have been widely utilized in magnetohydrodynamic (MHD) simulations in recent years. Due to the limited memory of a single GPU, distributed multi-GPU systems are needed to be explored for largescale MHD simulations. However, the data transfer between GPUs bottlenecks the efficiency of the simulations on such systems. In this paper we propose a novel GPU Direct-MPI hybrid approach to address this problem for overall performance enhancement. Our approach consists of two strategies: (1) We exploit GPU Direct 2.0 to speedup the data transfers between multiple GPUs in a single node and reduce the total number of message passing interface (MPI) communications; (2) We design Compute Unified Device Architecture (CUDA) kernels instead of using memory copy to speedup the fragmented data exchange in the three-dimensional (3D) decomposition. 3D decomposition is usually not preferable for distributed multi-GPU systems due to its low efficiency of the fragmented data exchange. Our approach has made a breakthrough to make 3D decomposition available on distributed multi-GPU systems. As a result, it can reduce the memory usage and computation time of each partition of the computational domain. Experiment results show twice the FLOPS comparing to common 2D decomposition MPI-only implementation method. The proposed approach has been developed in an efficient implementation for MHD simulations on distributed multi-GPU systems, called MGPU-MHD code. The code

show abstract

A new CUDA-based GPU implementation of the two-dimensional Athena code

Cited by 2 publications

References 12 publications

Bulletin of the Polish Academy of Sciences: Technical Sciences

Bulletin of the Polish Academy of Sciences: Technical Sciences

Efficient magnetohydrodynamic simulations on distributed multi-GPU systems using a novel GPU Direct–MPI hybrid approach

Contact Info

Product

Resources

About