Dr. Rao Mikkilineni: Cloud Computing through FCAPS Managed Services in a Virtualized Data Center.

Cloud Computing Conference - Cloud Slam 2009.
Abstract. With the advent of server, network and storage virtualization, applications can be dynamically provisioned to provide right CPU, memory, bandwidth, storage capacity and throughput and allocate appropriate resources based on workloads and business priorities. Such an end-to-end (application to spindle) virtualized data center provides a framework for next generation services creation, delivery and assurance platforms that are integrated to allow collaboration of hundreds of thousands of programmers to create millions of network based services used by billions of customers. The next generation cloud computing based Service Collaboration Network (SCN) will be analogous to the Intelligent Network (IN) in telecommunications that provides voice and data services. Service Assurance to both service developers and service end users is provided through end-to-end Fault, Configuration, Accounting, Performance and Security (FCAPS) management to allocate managed computing, network and storage virtualized resources on-demand to meet wide variations in business requirements and workload fluctuations.

For The SCN to be massively scalable and globally interoperable, this paper proposes a next generation Virtualized Resource Mediation Layer to provision CPU/memory, bandwidth and storage resources similar to the IN Services platform. In addition, the mediation layer must provide end to end FCAPS services to align the application requirements to available resources similar to the Telecommunications Management Network (TMN)

For the virtual resource mediation layer to be successful it has to be defined with a standards based RFI process and a standards body such as IETF or ITU should take leadership to drive global standards. The evolution of the telecommunications network and the Internet has demonstrated the success of this approach. By providing the mediation between virtualized computing, networking and storage resources, the VRML will become a Cloud Operating System (CLOS) and its domination by a single vendor can create monopoly that may not be in the best interest of the both consumers and suppliers. By learning from our past experiences with IN in telecommunications and IETF, we can accelerate the development of the CLOS and avoid single vendor monopolies.

Video of the session:

Assets:
http://www.kawaobjects.com/resources/kawaobjects-cloudslam09.pdf