A Tabu Search Algorithm for the Location of Data Centers and Software Components in Green Cloud Computing Networks
Issue No. 01 - January-June (2013 vol. 1)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TCC.2013.2
Federico Larumbe , Ecole Polytech. de Montreal & the Group for Res. in Decision Anal. (GERAD), Montreal, QC, Canada
Brunilde Sanso , Ecole Polytech. de Montreal & the Group for Res. in Decision Anal. (GERAD), Montreal, QC, Canada
The ubiquity of cloud applications requires the meticulous design of cloud networks with high quality of service, low costs, and low CO2 emissions. This paper presents a planning problem and an extremely efficient tabu search heuristic for optimizing the locations of cloud data centers and software components while simultaneously finding the information routing and network link capacities. The objectives are to optimize the network performance, the CO2 emissions, the capital expenditures (CAPEX), and the operational expenditures (OPEX). The problem is modeled using a mixed-integer programming model and solved with both an optimization solver and a tabu search heuristic. A case study of a web search engine is presented to explain and optimize the different aspects, showing how planners can use the model to direct the optimization and find the best solutions. The efficiency of the tabu search algorithm is presented for networks with up to 500 access nodes and 1,000 potential data center locations distributed around the globe.
Servers, Delays, Cloud computing, Power demand, Linear programming, Quality of service, Cloud computing
F. Larumbe and B. Sanso, "A Tabu Search Algorithm for the Location of Data Centers and Software Components in Green Cloud Computing Networks," in IEEE Transactions on Cloud Computing, vol. 1, no. 1, pp. 22-35, 2013.