Resource, data and application management for cloud federations and multi-clouds

• Autores: Vamis Xhagjika
• Directores de la Tesis: Leandro Navarro Moldes (dir. tes.), Vladimir Vlasov (codir. tes.)
• Lectura: En la Universitat Politècnica de Catalunya (UPC) ( España ) en 2017
• Idioma: español
• Tribunal Calificador de la Tesis: Rosa Maria Badia Sala (presid.), Joan Manuel Marques Puig (secret.), Paolo Trunfio (voc.)
• Programa de doctorado: Programa de Doctorado Erasmus Mundus en Computación Distribuida / Distributed Computing por la Universidad Politécnica de Catalunya; Instituto Superior Técnico de Lisboa(Portugal); Kungliga Tekniska Högskolan-The Royal Institute of Technology (Suecia) y Université Catholique de Louvain(Bélgica)
• Materias:
  • Matemáticas
    • Ciencia de los ordenadores
      • Informática
• Enlaces
  • Tesis en acceso abierto en: TDX
• Resumen

  • Distributed Real-Time Media Processing refers to classes of highly distributed, delay no-tolerant applications that account for the majority of the data traffic generated in the world today. Real-Time audio/video conferencing and live content streaming are of particular research interests as technology forecasts predict video traffic surpassing every other type of data traffic in the world in the near future. These applications are very sensitive to both communication properties such as latency, jitter, packet loss, bit rate as well as backend stream processing load profiles. In this work we provide a novel and generalized large-scale Multi-Cloud architectural blueprint for ISP and Carrier providers, that permits smart geo-distributed service placement in order to optimize latency/locality of stream processing applications. We provide as a well self-managed Intra-Cloud federation algorithm based on gradient topologies in order to optimize routes in a live media streaming backend. Additionally we introduce a novel distributed Network Bandwidth Manager that optimizes system stability by arbitrating network bandwidth between multiple Cloud services sharing the same network infrastructure. At last, an empirical study is provided connecting media quality parameters and Cloud backend load profiles, including an algorithm for stream allocation on Cloud Selective Forwarding units.