Exploring the future of cloud computing: Autonomous systems, edge artificial intelligence, and intelligent workload management

The field of cloud computing is rapidly evolving, with profound implications for organizations, consumers, and everyday life. While computing capabilities have been evolving for decades, users traditionally interacted with dispersed resources bound in the physical local area surrounding the user, or at least in their logical domain of knowledge. This computational island of services and resource capacity consisted of local and on-premise local clusters or enterprise data centers, hosting servers with their various servers and storage for shared use in delivering time-sharing or best-effort services to users in the domain. But, with the transformation of the internet into a globalization infrastructure, which is high bandwidth, low latency, and highly resilient, the key layers of the cloud computing stack became services on demand, renting resources for the long run without worrying about the underlying infrastructure. For example, a service allowed policies-based bulk storage for images, and another offered machines on demand that provided physical resources to run programs on a best-effort and time-shared basis (Gill et al., 2022; Ibn-Khedher et al., 2022; Mishra et al., 2024).

Having become a layer of modern internet operations, cloud computing is now an integral part of many, if not most, consumers' daily lives and business operations. A working group was initiated with the goal of producing Systems Engineering Guidance for Cloud Computing. But, cloud computing is complex, with many layers and issues involved; it is not always easy to know where to start, or what issues to consider when making one or more decisions about cloud computing. This chapter presents some background discussion of what cloud computing involves and some guidance for systems engineers, regardless of discipline or specialty area, who are faced with the decision or decisions about cloud computing.

Cloud computing is referred to as a huge pool of resources, which are not only shared by many users but also highly virtualized, so they can be conveniently accessed and used by heterogeneous end devices. Such devices would connect to the Internet via broadband wireless access since these huge pools of resources would be available worldwide. Data centers are physically distributed, and the only differences from general Internet servers are that they consist of a large number of machines with huge storage and broadband access to service a variety of user requests (Thota, 2024; Ramamoorthi, 2023; Walia et al., 2023).