Efficient skyline-based web service composition with QoS-awareness and budget constraint

In cloud computing, there exist several classes of web services, and each class of web service contains multiple services which have different Quality of Service (QoS) values. To provide QoS for users, determining the best service composition that meet user's preferences is necessity. Finding the best service composition needs a certain method in order to provide the best combination of QoS with minimum response time, while in some cases, the process is bounded by budget constraint provided in user's preferences. In a simple method, for selecting the best service composition, we need to evaluate n^m combinations of n classes of service and m services member. This condition leads to a large computing costs and memory consumption that occur during the process of generating the combination of data. To reduce the number of combinations, we propose an approach for finding the best service composition of web services using skyline algorithm which can adapt different user preferences of QoS. While selecting the best composition, we also take consideration in budget as constraint by selecting the closest user's budget with the best skyline result of services. SalSa skyline algorithm have given the new alternative ways to finding the best candidates of services in each class of service in reasonable time. These happened because we only need to evaluate the smaller parts of members in each class which will need much time if the number of services is plenty. Moreover, we also presented a method to find the best composition of services which able to present the composition that meet the user's budget.