Parallelizing Apriori on Dual Core using OpenMP

Accumulation of abundant data from different sources of the society but a little knowledge situation has lead to Knowledge Discovery from Databases or Data Mining. Data Mining techniques use the existing data and retrieve the useful knowledge from it which is not directly visible in the original data. As Data Mining algorithms deal with huge data, the primary concerns are how to store the data in the main memory at run time and how to improve the run time performance. Sequential algorithms cannot provide scalability, in terms of the data dimension, size, or runtime performance, for such large databases. Because the data sizes are increasing to a larger quantity, we must use high-performance parallel and distributed computing to get the advantage of more than one processor to handle these large quantities of data. The recent advancements in computer hardware for parallel processing is multi core or Chip Multiprocessor (CMP) systems. In this paper we present an efficient and easy technique for parallelization of apriori on dual-core using openMP wih perfect load balancing between the two cores. We present the performance evaluation of apriori for different support counts with different sized databases on dual core compared to sequential implementation.