This paper describes service clouds, a distributed infrastructure designed to facilitate rapid prototyping and deployment of adaptive communication services. The infrastructure combines adaptive middleware functionality with an overlay network substrate in order to support dynamic instantiation and reconfiguration of services. The service clouds architecture includes a collection of low-level facilities that can be invoked directly by applications or used to compose more complex services. After describing the service clouds architecture, we present results of experimental case studies conducted on the PlanetLab Internet testbed alone and a mobile computing testbed.

This paper describes Service Clouds, a distributed infrastructure designed to facilitate rapid prototyping and deployment of services that enhance communication performance, robustness, and security. The infrastructure combines adaptive middleware functionality with an overlay network substrate in order to support dynamic instantiation and reconfiguration of services. The Service Clouds architecture includes a collection of low-level facilities that can be either invoked directly by applications or used to compose more complex services. After describing the Service Clouds architecture, we present results of two experimental case studies conducted on the PlanetLab Internet testbed, the first to improve throughput of bulk data transfer, and the second to enhance the robustness of multimedia streaming

Peer-to-peer (P2P) anonymous communication systems are vulnerable to free-riders, peers that use the system while providing little or no service to others and whose presence limits the strength of anonymity as well as the efficiency of the system. Free-riding can be addressed by building explicit incentive mechanisms into system protocols to promote two distinct aspects of cooperation among peers-compliance with the protocol specification and the availability of peers to serve others. In this paper we study the use of payments to implement an incentive mechanism that attaches a real monetary cost to low availability. Through a game theoretic analysis, we evaluate the effectiveness of such an incentive, finding that peer availability can be significantly increased through the introduction of payments under many conditions. We also demonstrate how a payment-based incentive that preserves anonymity can be implemented and integrated with a popular class of P2P anonymity systems.

In pricing-based congestion control protocols it is common to assume that the rate of congestion feedback from the network is limited to a single bit per packet. To obtain a precise estimate of available bandwidth (as summarized by the congestion price) under the single-bit constraint, a session must consider feedback contained in a number of recently received packets. As more packets are considered, however, the estimate includes increasingly older information about the network state. We study this tradeoff between the quality and timeliness of feedback using control-theoretic approach, modeling the 'memory' incorporated into the price estimate as additional feedback delay. We show through analysis that obtaining arbitrary precision in the estimated price causes control instability, making it more difficult for a session to track its targeted optimal rate. Through continuous-time simulation of our model and packet-level simulations, we find that crude estimates of congestion price based on very few packets can yield good performance while allowing the session to operate far away from the boundary of instability. We also investigate the impact of estimation bias on protocol performance, showing that protocols use a form of integral control can compensate for biased price estimates.

One key component of recent pricing-based congestion control schemes is an algorithm for probabilistically setting the Explicit Congestion Notification bit at routers so that a receiver can estimate the sum of link congestion prices along a path. We consider two such algorithms - a well-known algorithm called Random Exponential Marking (REM) and a novel algorithm called Random Additive Marking (RAM). We show that if link prices are unbounded, a class of REM-like algorithms are the only ones possible. Unfortunately, REM computes a biased estimate of total price and requires setting a parameter for which no uniformly good choice exists in a network setting. However, we show that if prices can be bounded and therefore normalized, then there is an alternate class of feasible algorithms, of which RAM is representative and furthermore, only the REM-like and RAM-like classes are possible. For properly normalized link prices, RAM returns an optimal price estimate (in terms of mean squared error), outperforming REM even if the REM parameter is chosen optimally. RAM does not require setting a parameter like REM, but does require a router to know its position along the path taken by a packet. We present an implementation of RAM for the Internet that exploits the existing semantics of the time-to-live field in IP to provide the necessary path position information.