Search

Cloud computing has revolutionized service delivery by providing on-demand invocation and elasticity. To reap these benefits, computation has been displaced from client devices and into data centers. This partial centralization is undesirable for applications that have stringent locality requirements, e.g., low latency. This problem could be addressed with large numbers of smaller cloud resources closer to users. However, as cloud computing diffuses from within data centers and into the network, there will be a need for cloud resource allocation algorithms that operate on resource-constrained computational units that serve localized subsets of customers. In this paper, we present a mechanism for service provisioning in distributed clouds where applications compete for resources. The mechanism operates by enabling execution zones to assign resources based on Vickrey auctions and provides high-quality probabilistic models that applications can use to predict the outcomes of such auctions. This allows applications to use knowledge of the locality distribution of their clients to accurately select the number of bids to be sent to each execution zone and their value. The proposed mechanism is highly scalable, efficient, and validated by extensive simulations. [ABSTRACT FROM PUBLISHER]

This paper addresses the problem of provisioning management/monitoring nodes within highly dynamic network environments, particularly virtual networks. In a network, where nodes and links may be spontaneously created and destroyed (perhaps rapidly) there is a need for stable and responsive management and monitoring, which does not create a large load (in terms of traffic or processing) for the system. A subset of nodes has to be chosen for management/monitoring, each of which will manage a subset of the nodes in the network. A new, simple, and locally optimal greedy algorithm called Pressure is provided for choice of node position to minimize traffic. This algorithm is combined with a system for predicting the lifespan of nodes, and a tunable parameter is also given so that a system operator could express a preference for elected nodes to be chosen to reduce traffic, to be "stable,” or some compromise between these positions. The combined algorithm called PressureTime is lightweight and could be run in a distributed manner. The resulting algorithms are tested both in simulation and in a testbed environment of virtual routers. They perform well, both at reducing traffic and at choosing long lifespan nodes. [ABSTRACT FROM PUBLISHER]

We present a system for streaming live entertainment content over the Internet originating from a single source to a scalable number of consumers without resorting to centralized or provider-provisioned resources. The system creates a peer-to-peer overlay network, which attempts to optimize use of existing capacity to ensure quality of service, delivering low startup delay and lag in playout of the live content. There are three main aspects of our solution: first, a swarming mechanism that constructs an overlay topology for minimizing propagation delays from the source to end consumers; second, a distributed overlay anycast system that uses a location-based search algorithm for peers to quickly find the closest peers in a given stream; and finally, a novel incentive mechanism that encourages peers to donate capacity even when the user is not actively consuming content. [ABSTRACT FROM AUTHOR]