High-Performance Computing on Graphics Processing Units for Field and Device Modelling


Recent advances in multi-processor computer architectures provide many exciting opportunities in computational science and engineering. This is especially true for graphics processing units (GPU), which contain several hundred or thousands of stream processors and offer massive parallelization at a fraction of the cost and power consumption of comparable central processing unit (CPU) clusters. To further bolster their power, multiple GPUs can be installed in a single computer node or incorporated into multi-node GPU clusters, and used in heterogeneous CPU–GPU systems that allow researchers to exploit the strengths of both architectures. The use of GPUs already has led to major advances in many areas of scientific computing, including molecular dynamics, medical imaging, fluid dynamics, seismic imaging, and computational finance, to name a few. Needless to say, GPU computing also holds significant promise in field and device modelling.

There exist many challenges in porting present-day algorithms onto, or developing new algorithms for, GPUs. Indeed, GPU architectures differ significantly from those of CPUs and algorithms that execute efficiently on one of them may be ineffective on the other. The difficulties are further exacerbated when developing algorithms for heterogeneous GPU–CPU systems. Identifying and extending the range of applicability of GPU-accelerated or CPU–GPU hybrid methods is key to unlocking the full potential of this new computing paradigm.

The objective of this special issue is to report on recent advances in the area of GPU and hybrid CPU–GPU programming aimed at analyzing electrical engineering problems. The issue will include both invited and contributed papers. Topics of interest include, but are not limited to, theoretical developments with regard to computational methods suitable for implementation on GPUs, their practical implementation, and their use in the simulation of large-scale and complex field phenomena and the modelling of devices. Although the focus of the issue is on methods for modelling electronic networks, devices, and fields, contributions in broader areas related to GPU computing will be considered as well.

Manuscripts for this Special Issue should adhere to the requirements for regular papers of the IJNM as specified in the Author Guidelines at http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1099-1204/homepage/ForAuthors.html.

Potential contributors may contact the Guest Editor to determine the suitability of their contribution to the Special Issue. All manuscripts should be submitted via the IJNM's manuscript website http://mc.manuscriptcentral.com/ijnm, with a statement that they are intended for this Special Issue.

Manuscript submission deadline: 31 August 2011