Parallel Algorithm for Dynamic Data Fetch On Parallel Virtual Machine

摘要

Association Rules are used extensively for data mining in many domains such wholesale, retail marketing ,lottery selection and all the domains which is dealing with more number of data . Highly parallel algorithms for constructing association rules are desirable for dealing with large data sets in reasonable amount of time. Dynamic Programming merge patterns and mesh algorithms for building association rules have a natural concurrency, but are difficult to parallelize owing to the inherent dynamic nature of the computation. In this paper, we present parallel formulations of Dynamic Data Fetch Algorithm using Parallel Virtual Machine (PVM). We propose and describe five basic data parallel formulations to reduce the repetition of data. First one is Forward Path Cross Data Fetch(FPCDF) , Second is Reverse Path Cross Data Fetch(RPCDF), Third is Diagonal Cross Random Data Fetch(DCRDF), Fourth and fifth Upper/Lower Diagonal Cross Random Data Fetch(U/L DCRDF) and Data Merge andfetch(DMF) respectively.