Reliable Cloud Service Allocation based on Genetic Integration Under Multiple Parameters

In the public domain, multiple cloud services are available to satisfy user requirements. However, the automated effective selection of these services becomes more critical when the load on the server increases. In this paper, two-phase genetic optimization model is provided to achieve effective and reliable service allocation. The first level of genetic is applied on the server side to align the virtual machine based on resource availability and processing parameters. The fitness function and crossover operations are modified based on these parameters to generate the optimum VM sequence. The second level of genetic is applied on the client side based on requirement vectors. The request scheduling is obtained hereunder for the deadline criticality and resource requirement. Later on, the request sequence and availability sequence are mapped for effective and reliable service allocation. The implementation results show that the model has reduced the wait time and allocation failures. A decent reduction in request migration is also achieved from different experimentations.