This environment requires a network aware resource management which differs from resource management in localized datacenters. In order to meet applications’ requirements in terms of bandwidth and computing resources, one needs to consider link conditions and traffic engineering when allocating resources in a network embedded cloud. This paper presents a resource monitoring solution for a network embedded cloud by proposing an extension to the Open Shortest Path First-Traffic Engineering (OSPF-TE) protocols. Modeling, emulation, and analysis show the proposed solution can provide the required data to a cloud management system by sending information about virtual resources in a new opaque link-state advertisement, (cloud LSA). This cloud LSA is TE LSA which allows more efficient management of a network embedded cloud by providing available resource information (CPU, RAM, storage, and data center’s location) of each distributed data center as well as network conditions. Analysis shows in this solution each embedded data center injects less than 40 bytes per second of additional traffic into the network when sending cloud LSAs at most every 5 seconds.
Resource monitoring in a Network Embedded Cloud
A “network embedded cloud”, also known as a “carrier cloud”, is a distributed cloud platform where the computing resources are embedded in, and distributed across the carrier’s network. This is a transformation of carrier-grade networks toward a platform for cloud services and the emergence of cloud computing in a telecommunication environment. In this platform the network resources are simply another set of cloud resources.