GPU-Based Odd and Even Hybrid String Matching Algorithm

String matching is considered as one of the fundamental problems in computer science. Many computer applications provide the string matching utility for their users, and how fast one or more occurrences of a given pattern can be found in a text plays a prominent role in their user satisfaction. Although numerous algorithms and methods are available to solve the string matching problem, the remarkable increase in the amount of data which is produced and stored by modern computational devices demands researchers to find much more efficient ways for dealing with this issue. In this research, the Odd and Even (OE) hybrid string matching algorithm is redesigned to be executed on the Graphics Processing Unit (GPU), which can be utilized to reduce the burden of compute-intensive operations from the Central Processing Unit (CPU). In fact, capabilities of the GPU as a massively parallel processor are employed to enhance the performance of the existing hybrid string matching algorithms. Different types of data are used to evaluate the impact of parallelization and implementation of both algorithms on the GPU. Experimental results indicate that the performance of the hybrid string matching algorithms has been improved, and the speedup, which has been obtained, is considerable enough to suggest the GPU as the suitable platform for these hybrid string-matching algorithms.