Optimizing Costs for Cloud-Based Long-Running Activities using ILP

Abstract

Cloud infrastructure has increasingly emerged as a preferred solution for meeting business demands, particularly in executing various business processes, including long-running activities. Enterprises now faced with diverse cloud resources offered under quality-of-service (QoS)-based pricing strategies. As long-running activities within processes are time-sensitive, determining the ideal allocation of cloud resources becomes a complex task. This complexity arises from the need to fulfill activity quality standards, adhere to time constraints, and navigate through the complexities of cloud pricing strategies while optimizing resource costs. This paper presents an approach based on linear programming aimed at optimizing the allocation of cloud resources used in long-running activities. Our method's implementation and experimental results validate the feasibility of the proposed solution.