Apriori Algorithm Based Approach for Improving QoS and SLA Guarantee in IaaS Clouds Using Pattern-Based Service-Oriented Architecture

Cloud computing is a modern computing technology. The integration of fog computing, edge computing, mist computing, IoT, 5G, etc. with cloud computing adds immense power to cloud computing. However, this integration leads to two problems. The first problem is to define an interoperable cloud architecture and the second problem is resource allocation to optimize performance. To address both problems, we propose a SOA for cloud. We implement a machine learning based Apriori algorithm, to find association rules between QoS to define a stronger SLA. This approach also helps in defining better and simpler resource management policies and for developer it becomes easier to develop and deploy cloud applications. When cloud manages resources in a well-defined manner, revenue is optimized easily with maintaining required QoS levels. Experiments and detailed discussion show that this approach successfully helps in defining better SLA and implement a simpler resource management policy, makes easier to handle QoS, reduces costs and optimizes revenues.