A Robust Service Selection Method Based on Uncertain QoS

Abstract

Nowadays, the number of Web services on the Internet is quickly increasing. Meanwhile, different service providers offer numerous services with the similar functions. Quality of Service (QoS) has become an important factor used to select the most appropriate service for users. The most prominent QoS-based service selection models only take the certain attributes into account, which is an ideal assumption. In the real world, there are a large number of uncertain factors. In particular, at the runtime, QoS may become very poor or unacceptable. In order to solve the problem, a global service selection model based on uncertain QoS was proposed, including the corresponding normalization and aggregation functions, and then a robust optimization model adopted to transform the model. Experiment results show that the proposed method can effectively select services with high robustness and optimality.

Cloud manufacturing service composition and formal verification based on extended process calculus

“…On one hand, enterprise homogeneity is serious; on the other hand, due to the influence of geography, information, and other factors, the communication between enterprises or enterprise and customer is poor, resulting in ''isolated island phenomenon.'' 21 In view of the above problems, under the condition of limited manufacturing resources, in order to obtain the market competition advantage and win the maximum profit, machining enterprises need the support of the cloud manufacturing service composition platform, 22,23 thus to achieve the following: (1) improving the transparency and visibility of enterprises, quickly responding to market demand; (2) strengthening exchange and cooperation between enterprises in market, logistics, 24,25 product development, design, manufacturing, and so on; and (3) promoting the transformation of the industrial structure, optimizing the allocation of resources, and making full use of limited resources to get the maximum profit. 26 Xu et al 27 proposed a set of service domain-oriented artificial bee colony algorithms based on the optimization mechanism of artificial bee colony method for solving concurrent service selection and service composition problem.…”

Dynamic semantic‐based green bio‐inspired approach for optimizing energy and cloud services qualities

“…The first data set is a real data set collected by Zhang et al In this work, about 5000 cloud services were provided by 10 data centers. Currently, it has been applied in many studies . In this data set, and for each service, the values of its response time and throughput are varied according to its calls by the different users.…”

“…Currently, it has been applied in many studies. 17 In this data set, and for each service, the values of its response time and throughput are varied according to its calls by the different users. The second data set is a large random data set generated that consists in simulating 100 000 cloud services that represent generic (100) cloud services classified by domain (health, transport, and tourism), and for each generic service, the related candidate set of services contain (1000) specific services with different normalized QoS values.…”

“…Recently, Chen et al proposed dominance multiobjective evolutionary algorithm for solving the multiobjective QoSSSP. Chen et al have proposed an approach based on the nondominated sorting GA (NSGA‐II) to solve the multiobjective QoSSSP.…”

“…The QoS context‐aware and optimum resource allocation mechanisms have recently been proposed for reducing energy in data centers using QoS and context attributes as parameters. () However, these mechanisms do not address the changes of customer preferences (minimum costs, fast response time, high security, etc) during the discovery and the selection process of quality cloud services in the Cloud. Moreover, these techniques have not taken the dynamic nature of cloud system performance into account.…”