Solution of large complex problems in computational electromagnetics using higher order basis in MOM with parallel solvers
被引:0
作者:
Zhang, Y.
论文数: 0引用数: 0
h-index: 0
机构:
Syracuse Univ, Dept EECS, Syracuse, NY 13244 USASyracuse Univ, Dept EECS, Syracuse, NY 13244 USA
Zhang, Y.
[1
]
Sarkar, T. K.
论文数: 0引用数: 0
h-index: 0
机构:
Syracuse Univ, Dept EECS, Syracuse, NY 13244 USASyracuse Univ, Dept EECS, Syracuse, NY 13244 USA
Sarkar, T. K.
[1
]
Moon, H.
论文数: 0引用数: 0
h-index: 0
机构:
Syracuse Univ, Dept EECS, Syracuse, NY 13244 USASyracuse Univ, Dept EECS, Syracuse, NY 13244 USA
Moon, H.
[1
]
De, A.
论文数: 0引用数: 0
h-index: 0
机构:
Syracuse Univ, Dept EECS, Syracuse, NY 13244 USASyracuse Univ, Dept EECS, Syracuse, NY 13244 USA
De, A.
[1
]
Taylor, M. C.
论文数: 0引用数: 0
h-index: 0
机构:
Syracuse Univ, Dept EECS, Syracuse, NY 13244 USASyracuse Univ, Dept EECS, Syracuse, NY 13244 USA
Taylor, M. C.
[1
]
机构:
[1] Syracuse Univ, Dept EECS, Syracuse, NY 13244 USA
来源:
2007 IEEE ANTENNAS AND PROPAGATION SOCIETY INTERNATIONAL SYMPOSIUM, VOLS 1-12
|
2007年
关键词:
computational electromagnetics;
method of moments;
Vivaldi array;
electrically large objects;
parallel computation;
D O I:
暂无
中图分类号:
TM [电工技术];
TN [电子技术、通信技术];
学科分类号:
0808 ;
0809 ;
摘要:
In this paper, a general purpose parallel electromagnetic (EM) solver is presented for complicated EM scattering and radiation problems. The parallel algorithm has been designed for multiple platforms and is presented here on four typical operating systems with different hardware, which include IA32 with Windows, IA64 with Windows and EM64T with Linux and Windows operating systems and a DELL PowerEdge 1855 Blade Cluster and Itanium II server in hardware. Techniques have been successfully developed to solve the problem of access to more than 2 GB RAM per process. Since there is no commercial parallel ScaLAPACK solver available for Itanium 2 with Windows Operating System, we also developed ScaLAPACK solver which is presented in this paper. Numerical results shown in this paper indicate that the parallel solver can be run across multiple platforms to solve complicated EM problems very efficiently.