ICSD: Integrated Cloud Services Dataset

Abstract

The service composition problem in Cloud computing is formulated as a multiple criteria decision making problem. Due to the extensive search space, Cloud service composition is addressed as an NP-hard problem. Using a proper dataset is considered one of the main challenges to evaluate the efficiency of the developed service composition algorithms. According to the work in this paper, a new dataset has been introduced, called Integrated Cloud Services Dataset (ICSD). This dataset is constructed by amalgamating the Google cluster-usage traces, and a real QoS dataset. To evaluate the efficiency of the ICSD dataset, a proof of concept has been done by implementing and evaluating an existing Cloud service compositing approach; PSO algorithm with skyline operator using ICSD dataset. According to the implementation results, it is found that the ICSD dataset achieved a high degree of optimality with low time complexity, which significantly increases the ICSD dataset accuracy in Cloud services composition environment.