The Case for Generic Edge Based Services

• Autor(a) principal: Jesus, Diogo Alexandre Matos de
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/163259
• topic: Edge Computing; Causal Consistency; Fault Tolerance; Security; Replication; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

In the present day, most, if not all, applications or services are available worldwide. In practice, developers achieve this by replicating the state of a system in different parts of the globe, in a centralized manner, through data centers. With the rapid growth of devices connected to the internet and the amount of applications running on them, traditional centralized computing models, in which data is sent to data centers for processing and storage, are becoming increasingly costly and inefficient. Edge computing allows to store and process data close to the users, at the network’s periphery, reducing the amount of data that needs to be sent to centralized data centers, while allowing shorter response times to the user. Applications can benefit from the functionality provided in the edge, from caching to telemetry and privacy guarantees, by allowing computations on these access points. Deploying independent services for each application is costly and impractical, making generic services an alternative approach. Several parties (i.e., cloud platforms) are expanding their infrastructures to the edge to put computation closer to the end user and allow faster response times. New technologies such as 5G networks and the increase of computing nodes in points of presence next to the user, keep increasing the options and platforms for running computations on the edge and the feasibility of services running in this paradigm at a large scale. In this work, we study how to use and implement generic services deployed at the network’s edge, in order to enhance the development of decentralized applications and explore the gains and benefits of placing computing close to the clients. In particular, we focus on services that can enhance the replication in decentralized systems, such as multiplayer games, collaborative applications, and secure blockchain-based applications.