We present a performance analysis of a parallel implementation for both preconditioned conjugate gradient and preconditioned bi-conjugate gradient solvers running on graphic processing units (GPUs) with CUDA programming model. The solvers were mainly optimized for the solution of sparse systems of algebraic equations at complex entries, arising from the three-dimensional edge-finite element analysis of the electromagnetic phenomena involved in the open-bound earth diffusion of currents under time-harmonic excitation. We used a shifted incomplete Cholesky (IC) factorization as preconditioner. Results show a significant speedup by using either a single-GPU or a multi-GPU device, compared to a serial central processing unit (CPU) implementation, thereby allowing the simulations of large-scale problems in low-cost personal computers. Additional experiments of the optimized solvers show that its use can be extended successfully to other complex systems of equations arising in electrical engineering, such as those obtained in power–system analysis.
AmentMGKnittelDWeiskopfW. (2010) A parallel preconditioned conjugate gradient solver for the Poisson problem on a multi-GPU platform. In: IEEE 18th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), Pisa, Italy, February, pp. 583–592. IEEE.
2.
BarretRBerryMChanTF. (1994) Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods. 2nd ed. Philadelphia, PA: SIAM.
CamargosAFPSilvaVCGuichonJ-M. (2014a) Efficient parallel preconditioned conjugate gradient solver on GPU for FE modeling of electromagnetic fields in highly dissipative media. IEEE Transactions on Magnetics50(2): 569–572.
5.
CamargosAFPSilvaVCGuichonJ-M. (2014b) Iterative solution on GPU of linear systems arising from the A-V edge-FEA of time-harmonic electromagnetic phenomena. IEEE 22nd Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), Turin, Italy, February, pp. 365–371. IEEE.
6.
FujiwaraKNakataTFusayasuT (1993) Acceleration of convergence characteristic of the ICCG Method. IEEE Transactions on Magnetics29(2): 1958–1961.
7.
HelfensteinRKokoJ (2012) Parallel preconditioned conjugate gradient algorithm on GPU. Journal of Computational and Applied Mathematics236(15): 3584–3590.
8.
HestenesMRStiefelE (1952) Methods of conjugate gradients for solving linear systems. Journal of Research of the National Bureau of Standards,49(6): 409–436.
9.
IdemaRPapaefthymiouGLahayeD. (2013) Towards faster solution of large power flow problems. IEEE Transactions on Power Systems28(4): 4918–4925.
10.
IkunoSKawaguchiYFujitaN. (2012) Iterative solver for linear system obtained by edge element: Variable preconditioned method with mixed precision on GPU. IEEE Transactions on Magnetics48: 467–470.
11.
KamiabadAATateJE (2013) Polynomial preconditioning of power system matrices with graphics processing units. In: High Performance Computing in Power and Energy Systems. Heidelberg: Springer Berlin, pp. 229–246.
12.
LiRSaadY (n.d.) GPU-accelerated preconditioned iterative linear solvers. Available at: http://citeseerx.ist.psu.edu/index (accessed April 2015).
LumsdanieASiekJ (2014) ITL: The Iterative Template Library. Available at: www.osl.iu.edu/research/itl (accessed May 2015).
15.
MartinhoLBSilvaVCFilhoMLP. (2014) 3D finite element analysis of conductive coupling problems in transmission line rights of way. IEEE Transactions on Magnetics50(2): 969–972.
16.
MesquitaRC (1994) Additional properties of the incomplete gauge formulation for 3-D nodal finite-element magnetostatics. IEEE Transactions on Magnetics30(5): 2861–2864.
17.
MoghniehHLowtherDA (2010) Understanding the efficiency of parallel incomplete Cholesky preconditioners on the performance of ICCG solvers for multi-core and GPU systems. IEEE 14th Biennial Conference on Electromagnetic Field Computation (CEFC), Chicago, IL, USA, January. IEEE.
18.
NaumovM (2011) Incomplete-LU and Cholesky preconditioned iterative methods using CUSPARSE and CUBLAS. NVIDIA Corporation. Available at: https://developer.nvidia.com (accessed June 2011).
19.
OwensJDHoustonMLuebkeD. (2008) GPU computing. Proceedings of IEEE96(5): 879–899.
20.
RodriguesAWOGuyomarc’hFDekeyserJ-L. (2012) Automatic multi-GPU code generation applied to simulation of electrical machines. IEEE Transactions on Magnetics48(2): 831–834.
21.
SaadY (2003) Iterative Methods for Sparse Linear Systems,2nd edn. New York: PWS Publishing.
22.
ShewchukJR (1994) An introduction to the conjugate gradient method without the agonizing pain. Technical report, Carnegie Mellon University, Pittsburgh, PA, USA.
23.
SilvaVCMartinhoLBCardosoJR. (2011) Efficient modeling of thin wires in a lossy medium by finite elements applied to grounding systems. IEEE Transactions on Magnetics47(5): 966–969.
24.
StupfelB (2004) A Study of the condition number of various finite element matrices involved in the numerical solution of Maxwell’s equations. IEEE Antennas Propagation52(11): 3048–3059.
25.
Van der SluisAVan der VorstHA (1986) The rate of convergence of conjugate gradients. Numerische Mathematik48(5): 543–560.
