Accelerated EMF Evaluation Using a SIMD Algorithm

´´ ´ ´´ ´ ´´ ´ Abstract. This article presents a fast Single-Instruction Multiple-Data (SIMD) algorithm that evaluates electromagnetic fields (EMFs). It is based on the Method of Moments (MOM) adapted for execution on an SIMD architecture. The big speed-up obtained with this new implementation enables us to obtain results faster and to simulate more complex and realistic models and keep the computing time within a reasonable range, which give lead to better solutions for current EMF problems. This article gives a brief overview of generic massively parallel processors, taking into consideration their hardware architecture and the new computer languages for managing them. We describe the mathematical foundations of the algorithm in order to explain how the operations are distributed and performed by the GPU. Many cases are simulated to analyze the performance of the method proposed and they are compared with a fully implemented CPU algorithm, as well as with another CPU algorithm that uses the Intel MKL solvers for dense matrices. The differences in performance between floating-point precision numbers and double precision numbers is also studied and how they influence the accuracy of the results. The tests carried out suggest that the acceleration obtained grows with the complexity of the model. As a result, the proposed algorithm's only limitations lie with the hardware features. Streszczenie. Artykul przedstawia szybki algorytm typu Single-Instruction Multiple-Data (SIMD - pojedyncza instrukcja wiele danych), do obliczania rozkladu pol elektromagnetycznych (EMF). Jest ona oparta na metodzie momentow (MOM) przystosowanych do realizacji w architekturze SIMD. Du˙ ze przyspieszenie uzyskane dzi ˛ tej nowej implementacji pozwala uzyskac wyniki szybciej i dla bardziej zlo˙ zonych symulacji i realistycznych modeli i utrzymac czas obliczeniowy w rozs ˛ adnym zakresie. Artykul zawiera krotki przegl ˛ ad procesorow masowo rownoleglych, bior ˛ ac pod uwag ˛ e ich architek- tur ˛ e sprz˛ etow ˛ ain owe j ˛ ezyki programowania do zarz ˛ adzania nimi. Opisano matematyczne podstawy algorytmu, aby wyjasnic, w jaki sposob operacje s˛ a wykonywane przez procesor graficzny (GPU). Wiele przypadkow zostalo symulowanych aby przeanalizowac dzialanie proponowanej metody a wyniki porownano ze znanymi algorytmami, jak rownie˙ z z algorytmem, ktory wykorzystuje Intel MKL Solvers do g ˛ estych matryc. Z przeprowadzonych testow wynika, ˙ uzyskane przyspieszenie rosnie wraz ze zlo˙ zonosci ˛ a modelu. Jedyne ograniczenia algorytmu zale˙ z˛ ao d mosprz˛ etowych. (Przyspieszony algorytm analizy pol elektromagnetycznych za pomoc ˛ a algorytmu SIMD)

[1]  Jie Cheng,et al.  CUDA by Example: An Introduction to General-Purpose GPU Programming , 2010, Scalable Comput. Pract. Exp..

[2]  T. Sarkar,et al.  Analysis of Multiconductor Transmission Lines , 1988, 31st ARFTG Conference Digest.

[3]  C. Vilacha,et al.  Magnetic-Field Evaluation in the Vicinity of High-Voltage Electric Systems , 2012, IEEE Transactions on Power Delivery.

[4]  T Topa,et al.  Using GPU With CUDA to Accelerate MoM-Based Electromagnetic Simulation of Wire-Grid Models , 2011, IEEE Antennas and Wireless Propagation Letters.

[5]  Venkata Dinavahi,et al.  SIMD-Based Large-Scale Transient Stability Simulation on the Graphics Processing Unit , 2010, IEEE Transactions on Power Systems.

[6]  Mladen Kezunovic,et al.  The 21st century substation design: Vision of the future , 2010, 2010 IREP Symposium Bulk Power System Dynamics and Control - VIII (IREP).

[7]  Hubert Nguyen,et al.  GPU Gems 3 , 2007 .

[8]  Tomasz Topa,et al.  Adapting MoM With RWG Basis Functions to GPU Technology Using CUDA , 2011, IEEE Antennas and Wireless Propagation Letters.

[9]  Markus Clemens,et al.  Scalability of Higher-Order Discontinuous Galerkin FEM Computations for Solving Electromagnetic Wave Propagation Problems on GPU Clusters , 2010, IEEE Transactions on Magnetics.

[10]  A. F. Otero,et al.  Nodal frequency analysis of grounding systems considering the soil ionization effect , 2000 .

[11]  Taskin Koçak,et al.  Smart Grid Technologies: Communication Technologies and Standards , 2011, IEEE Transactions on Industrial Informatics.

[12]  A. F. Otero,et al.  Frequency-dependent grounding system calculation by means of a conventional nodal analysis technique , 1999 .

[13]  Soichi Watanabe,et al.  A GPU-based calculation using the three-dimensional FDTD method for electromagnetic field analysis , 2010, 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology.

[14]  Jie Cheng,et al.  Programming Massively Parallel Processors. A Hands-on Approach , 2010, Scalable Comput. Pract. Exp..

[15]  Eric J. Kelmelis,et al.  CULA: hybrid GPU accelerated linear algebra routines , 2010, Defense + Commercial Sensing.

[16]  H.-D. Bruns,et al.  Numerical Electromagnetic Field Analysis for EMC Problems , 2007, IEEE Transactions on Electromagnetic Compatibility.

[17]  Jon Rigelsford,et al.  Laws and Models: Science, Engineering and Technology , 2003 .

[18]  Alex Fit-Florea,et al.  Precision and Performance: Floating Point and IEEE 754 Compliance for NVIDIA GPUs , 2011 .