Automated decision-making for scaling a cloud application
Tamara Savchuk , Andriy KozachukArticle describes the algorithm of automatic decision making on scaling of a cloud application based on a set of rules of reactive scaling and on a function that estimates each possible way of scaling of a cloud application using calculation of lost user count and price of cloud hosting equipment. Article introduces evaluation function of scaling options that allows obtaining economic evaluation of the efficiency based on the cost of maintaining infrastructure in the hosting cloud and estimation of the number of users that stop using cloud applications due to overheating-intensity. The forecast number of orphaned users is made on the basis of calculating the network requests execution time. Developed algorithm can be used as part of a cloud application scaling technology