Development of a Model using Memory based Parallel Algorithm for Continue Matrix Multiplication

Chain multiplication of matrices is wide used for scientific computing. It becomes more difficult once there is sizable amount of floating purpose dense matrices. Because, floating purpose operations take longer than whole number operations. It could be attention-grabbing to lower the time of such chain operations. Now-a-days each multicore processor system has in-built parallel machine power. This power will solely be utilised once compatible parallel algorithms were used. So, during this work, a shared memory primarily based parallel algorithms has been projected to cipher the multiplication of a protracted sequence of dense matrices. The algorithms are tested with long sequence of matrices as input. The approach has been with 2×108 flops. The input matrix sequence length was generally varied from two to thirty. Most range of processors used was eight (Eight core processor).Different parameters like quickening, potency etc. were additionally noted. It had been all over that .The parallel algorithms might win roughly ninetieth potency at the best case. The algorithms additionally showed improved measurability.