A cloud-based optimization for the placement of water quality sensors in water distribution system

Optimal placement of the water quality sensor refers to the arrangement of best water quality sensor deployment in water distribution systems to test the accidental or intentional contaminant. The optimization problem is essentially a kind of large scale combinatorial optimization problem and it needs the highly reliable and highly scalable resources to solve the problem. Traditional methods to solve this kind of problems are usually using MPI. But it can´t meet the demand of QoS (quality of service) which contains reliability, scalability and flexibility. In order to solve this problem, we have developed a solution which is parallel and we use the Map/Reduce computation model in Clouds, namely the parallel GA. It uses Map/Reduce framework to parallelize the process. A middle-sized water distribution network is tested by the proposed algorithm, and the results indicate that the parallel GA with Map/Reduce can cut the computing time and deliver a high performance.