Your location now is: Home > ACADEMIC> Content

A Hybrid Particle Swarm Ant Colony Based Resource Reservation for Geo-distributed Cloud Service

十二月 13,2016 Click:

Conference name: IEEE-GLOBECOM2016

IEEE Global Communications Conference (IEEE GLOBAL COMMUNICATIONS CONFERENCE)

Report time: December 13, 2016 (Tuesday) at 15:00 PM

Report location: Railway Campus second comprehensive experimental building, 513

Speaker: Song Yazhen

Report title:A Hybrid Particle Swarm Ant Colony Based Resource Reservation for Geo-distributed Cloud Service

Report summary:Abstract—In cloud market, cloud providers offer diverseservice options, including on-demand instances and reservedinstances. Generally reserved instance price is cheaper thanon-demand instance, but excessive reservation may result inhigh capacity underutilization. To improve resource utilizationrate and reduce providers’ service cost, a steady broker federationis necessary to propose. Broker federation coordinatecloud providers service demands in geo-distributed data centersthrough deciding when and how many instances to reserve.Firstly, the service cost optimization problem is formulatedas a nonlinear integer programming model. Then a hybridalgorithm combining ant colony with particle swarm optimizationis presented to reduce computational complexity and providersservice cost. Extensive simulations driven by large-scale ParallelWorkloads Archive demonstrate the effectiveness and efficiencyof the hybrid algorithm.


Last : Resource Allocation for Cloud-Radio Access Network Powered by Renewable Energy

Next : Cryptography in a quantum world