采用有限元法進行電法勘探時，會產生大型稀疏線性方程組，如何提高方程組的求解效率成為物探研究的關鍵。針對傳統直接法難以實現并行求解的缺點，提出了在Beowulf 集群環境下，采用并行PCG 算法求解物探系統線性方程組。在集群環境下，該算法具有機器間相互通訊少、時間復雜度低等優點，并且易于并行實現。實驗結果表明，采用PCG 算法獲得了良好的并行效果。
關鍵詞：有限元法；Beowulf 集群；預處理共軛梯度法；并行虛擬機
Research on Application of Parallel PCG Algorithm in Electrical Surveying CHEN Rong-zheng1，LI Dai-ping1，HE Chi2,HUANG Jian1 (1：Computer Department of Guangdong University of Technology；2：International Trade and Economic Department of Guangdong University of Foreign Studies) Abstract：The sparse sets of linear equations are produced in electrical surveying using finite element method，how to raise the efficiency of the solution of equations is the key to Object-probed.In order to overcome the demerit that traditional direct method is difficult to parallelize, PCG algorithm based on Beowulf cluster is put
forward to solve linear equations.The algorithm has the merits of less communication ,
lower time complexity and parallel implementation is also very easy under the
environment of Beowulf Cluster. Experimental results show that good parallel results
have been obtained using the algorithm.
Key words：finite element method；Beowulf Cluster； preconditioned conjugate gradient
method；PVM