Concurrent Differential Evolution Based on MapReduce
Source: NORTH ATLANTIC UNIVERSITY UNION
Multi-core processors, which have more than one Central Processing Unit (CPU), have been introduced widely into personal computers. Therefore, in order to utilize the additional cores, or CPUs, to execute various costly application programs, concurrent implementations of them have been paid to attention. MapReduce is a concurrent programming model and an associated implementation for processing and generating large data sets. This paper has been participated in plenary presentation at the conference of WSEAS and is presenting a further progress of a concurrent implementation of Differential Evolution (DE) based on MapReduce. Especially, through the numerical experiment conducted on a wide range of benchmark problems, the speedup of DE due to the use of multiple cores is demonstrated.