Research On Collaborative Provisioning-oriented Cloud Service Selection Mechanism

Cloud computing which is based on Internet technology, integrates all kinds of resources in the network to form a distributed service cluster system, and provides the users with computational resources which reduced or expanded in a virtualized way. The goal of cloud computing is to provide all kinds of online services for lots of users and the terminals. In the cloud service, the key problem users concerned about is that the services they can get, instead of computational resources. In the cloud computing environment, the combination of distributed resources on demand to achieve between users and service providers dynamic, flexible and effective autonomous cooperative, and the cloud service selection mechanism is formed on this basis, the mechanism is becoming an important factor which influences and promotes the cloud computing technology and industry development. At present, the cloud service selection methods are mostly in view of the static service resource, lack of dynamic and ignoring the inherent uncertainty of Qo S; In addition, a large number of service redundancy leads to poor real-time cloud service selection; Because of the problems that are the users of Internet are larger and the users preference personality, the traditional service selection methods cannot be in good application in the cloud environment.In this paper, we concentrate on the efficiency, the real-time problem and other issues of cloud service selection. Based on the analysis of existing cloud service selection technologies, we converge Chord ring to improve the efficiency of cloud service selection. And increasing the fuzzy weight of user preference to make implementation of uncertainty, dynamic changing needs of user preference. Finally we provide a combination of cloud services fast and globally for the user through bacterial foraging optimization algorithm. The main research work includes the following aspects:(1)The existing cloud service selection key technologies were analyzed, we comprehensively and systematically exposit the current research progress. The main problems existing in current cloud service selection were introduced, and the current research methods related cloud service technologies were discussed.(2)Selection method is proposed based on fuzzy Qo S cloud service, aiming at efficient cloud service selection and the weight uncertainty of user preference. Using this method, on one hand, a virtual cloud service resource integration model were proposed based on Chord which can make an efficient retrieval of candidate services according to the need of users. On the other hand, because of the weight uncertainty of user preference, Fuzzy Analytic Hierarchy Process is introduced to calculate the user preference weight which can make a more comprehensive, more accurate representation of weight.(3)When a single cloud service cannot meet user needs, new cloud service composition method based on bacterial foraging optimization algorithm is proposed. Cloud service composition problem is a typical NP-hard problem, the bacterial foraging optimization algorithm has many advantages, such as intelligent search, easily jump out of local minimum value etc. Therefore, in order to make search service effectively-quickly and combine them together to provide the user, bacterial foraging optimization algorithm is introduced to the cloud service composition optimization.