End-to-End Service Quality for Cloud Applications

This paper aims to highlight the importance of end-to-end service quality for cloud services, with a focus on telecom carrier-grade services. In multi-tenant distributed and virtualized cloud infrastructures, the enhanced level of resource sharing raises issues in terms of performance stability and reliability of cloud services, threatening the possibility to offer precise service levels in end-to-end scenarios. Technology-wise, we have today some basic building blocks that may enable cloud infrastructures to exhibit stable and predictable performance to customers. However, one of the major obstacles that keep hindering the potential for a worldwide deployment of these technologies, is the fact that, in many distributed and cloud computing scenarios, there is not merely a single business entity responsible for the service delivery, but we may have multiple different, unrelated business entities with contrasting and competing requirements, interacting for the provisioning of end-to-end cloud services to customers and finally end users. It is the case, for example, of multiple cloud providers, storage providers and network service providers, that may be involved for the delivery of a distributed cloud service to a community of end users. In this context, setting up proper Service-Level Agreements (SLAs) among the involved players, for delivering strong QoS guarantees to customers, may become overly challenging. However, the main problems arising in such interactions may be mitigated by a thoughtful intervention of standardization. The paper reviews some of the most important efforts in research and industry to tackle end-to-end service quality and concludes with some recommendations for additional required research and/or standardization effort required to be able to deploy mission critical or interactive real-time services with high demands on service quality, reliability and predictability on cloud platforms.