发现信息

  • 作者:Gaetano Francesco Pittalà
  • 年份:2026
  • 发表源:AMS Dottorato Institutional Doctoral Theses Repository (University of Bologna)
  • OpenAlex ID:https://openalex.org/W7225554550
  • DOI:https://doi.org/10.48676/unibo/amsdottorato/13294
  • arXiv:待核验
  • 检索主题:agent-memory
  • OpenAlex 引用数快照:0(2026-11-12)

摘要线索

Edge computing has become an important way to support next-generation telecommunications networks. By moving computation closer to data sources and end users, it makes it possible to provide low-latency, energy-efficient, and context-aware services. This dissertation examines the orchestration and scheduling of heterogeneous services across various edge computing architectures, tackling the inherent complexity arising from the diversity of network infrastructures, computing nodes, and service requirements. The first part of the thesis is about how to design and analyze an edge-oriented service orchestration platform. This platform is designed to handle different types of services over different types of networks, taking into account the unique features of edge environments, like limited resources, architectural diversity, and changing operating conditions. The research's main focus is on service scheduling strategies, with a focus on optimizing key performance indicators that are important for future telecommunications networks, such as end-to-end latency and energy use. Various scheduling methodologies are proposed and assessed to concurrently optimize these objectives, emphasizing the trade-offs among performance, resource utilization, and operational efficiency in distributed edge contexts. The thesis further investigates the implementation of these mechanisms in scenarios where the edge nodes' interconnection is not stable, introducing possible additional mechanisms to deal with such uncertainties of the communication channels. Following, the focus shifts towards the examination of security-oriented scenarios in industrial settings, highlighting how the utilization of diverse computational capabilities within the network can improve both performance and resilience. Lastly, this manuscript explores the application of the edge computing model to optical networking settings, looking into new ways to make optical resources, like wavelength deployment, available as a service. This final part, influenced by cloud computing principles, examines the abstraction and orchestration of optical network components in a service-oriented framework, facilitating flexible, on-demand provisioning and enhanced integration between computing and networking resources.

核验与筛选

  • 当前状态:待筛选、去重和一手来源核验。
  • OpenAlex 元数据和引用数只用于发现,不作为正式引文或质量判断。
  • 记录正式版本、venue、方法细节、实验协议和局限后,才迁入稳定论文卡片。