Thumbnail Image

Optimization of Elastic Cloud Brokerage Mechanisms for Future Telecommunication Service Environments

Campowsky, Konrad; Carella, Giuseppe; Magedanz, Thomas; Schreiner, Florian

FG Architekturen der Vermittlungsknoten

Cloud computing mechanisms and cloud-based services are currently revolutionizing Web as well as telecommunication service platforms and service offerings. Apart from providing infrastructures, platforms and software as a service, mechanism for dynamic allocation of compute and storage resources on-demand, commonly termed as “elastic cloud computing” account for the most important cloud computing functionalities. Resource elasticity allows not only for efficient internal compute and storage resource consumption, but also, through so called hybrid cloud computing mechanisms, for dynamic utilization of external resources on-demand. This capability is especially useful in order to cost-efficiently cope with peakworkloads, allowing service providers to significantly reduce usually required over-provisioned service infrastructures, allowing for “pay-per-use” cost models. With a steadily growing number of cloud providers and with the proliferation of unified cloud computing interfaces, service providers are given free choice of flexibly selecting and utilizing cloud resources from different cloud providers. Cloud brokering systems allow for dynamic selection and utilization of cloud computing resources based on functional (e.g. QoS, SLA, energy consumption) as well as nonfunctional criteria (e.g. costs). The presented work focuses on enhanced cloud brokering mechanisms for telecommunication service platforms, enabling quality telecommunication service assurance, still optimizing cloud resources consumption, i.e. saving costs and energy. Furthermore this work shows that by combining cloud brokering mechanisms with standardized telecommunication service brokering mechanisms an even greater benefit for telecommunication service providers can be achieved as this enables an even better cost-efficiency since different user segments can seamlessly be served by allocating different cloud resources to them in a policy-driven manner.
  • Dieser Beitrag ist mit Zustimmung des Rechteinhabers aufgrund einer (DFG geförderten) Allianz- bzw. Nationallizenz frei zugänglich.
  • This publication is with permission of the rights owner freely accessible due to an Alliance licence and a national licence (funded by the DFG, German Research Foundation) respectively.