An Active Service Framework and its Application to Real time Multimedia Transcoding Elan Amil StevenMcCanne and Randy Katz University of California Berkeley elan mccanne randy EECS Berkeley EDU Abstract 1 Introduction One of the key strengths of the Internet service model is that it abstracts away the details of how messages are forwarded through the network On the one hand this design principle is extremely powerful because it divorces applications from much of the complexity of the underlying communication system but at the same time it constrains applications because they cannot exploit detailed knowledge of the underlying network to enhance their performance One application class that is constrained in this way are so called video gateways 3 which are computational elements that adjust the bit rate of video stream or collection of video streams to accommodate the constrained capacity of communication links at strategic locations within the network Becausevideo gateways perform their optimization directly on the underlying data stream from within the network itself they must be physically situated at an appropriate perhaps arbitrary point within the network However even though these agents are deeply embedded within the network infrastructure they are actually created configured and controlled dynamically by the user application on the end system at the edge of the network for instance by employing application specific protocols to convey receiver interest into the network to best configure the agent for the receiving user s preferences and capabilities 2 Unfortunately the Internet service model has no native support for deploying agents within the network in this fashion To overcome this limitation the Active Networks initiative 30 proposes that the Internet service model be replaced with a new architecture in which the network as a whole becomes a fully programmable computational environment In this framework not only do application defined entities run on arbitrary nodes in the network but individual packets can be programmed to perform arbitrary actions as they propagate through the network programmability migrates down the protocol stack from the application layer across the transport layer and into the network and datalink layers While the requirements of video gateways are often cited in the active networks literature as proof of a pressing need for this new infrastructure the far reaching implications of uprooting and supplanting over twenty years of Internet design experience begs the question Is active networks both sufJicient and necessary for deploying scalable flexible and robust services like the video gateway service within the network on behalf of the user While we concede that a comprehensive active networks framework would immediately solve this problem and is therefore sufficient we argue that for a large and important sub class of the active networks design space namely those applications like the video gateway that provide an application level service the active networks architecture is not strictly necessary Instead we argue that these applications are adequately and effectively supported by a programmable service architecture built on top of and in harmony with the existing Internet service model that allows users to download and run code on their behalf at strategic locations within the Several recent proposals for an active networks architecture advocate the placement of user defined computation within the network as a key mechanism to enable a wide range of new applications and protocols including reliable multicast transports mechanisms to foil denial of service attacks intra network real time signal transcoding and so forth This laudable goal however creates a number of very difficult research problems and although a number of pioneering research efforts in active networks have solved some of the preliminary small scale problems a large number of wide open problems remain In this paper we propose an alternative to active networks that addresses a restricted and more tractable subset of the active networks design space Our approach which we and others call active services advocates the placement of userdefined computation within the network as with active networks but unlike active networks preserves all of the routing and forwarding semantics of current Internet architecture by restricting the computation environment to the application layer Because active services do not require changes to the Internet architecture they can be deployed incrementally in today s Internet We believe that many of the applications and protocols targeted by the active networks initiative can be solved with active services and toward this end we propose herein a specific architecture for an active service and develop one such service in detail the Media Gateway MeGa service that exploits this architecture In defining our active service we encountered six key problems service location service control service management service attachment service composition and the definition of the service environment and have crafted solutions for these problems in the context of the MeGa service To verify our design we implemented and fielded MeGa on the UC Berkeley campus where it has been used regularly for several months by real users who connect via ISDN to an on line classroom Our initial experience indicates that our active services prototype provides a very flexible and programmable platform for intra network computation that strikes a good balance between the flexibility of the active networks architecture and the practical constraints of incremental deployment in the current Internet Permission to make digital or herd copies of alI or PM 0f this work for personas or ckwoom use is Sranted without fee provided that copies we not made or distributed for proflt or co rCia adVanteSe and that copses bear this notice and the full citation on the fwt Page To copy otherwie e to republish to post on server8 Or to redistribute to lusts requlrss prior epaciflc permission and Or B fee SlGCOMM 98 Vancouvsr B C 0 998 ACM l 58113 003 1 98KKx S 5 178 network We and others call this programmable network infrastructure active services because we restrict the design to applicationlevel services yet we inherit the novelty of the active networks programmability 161 While we believe that the active services framework holds the promise to enable many active networks like applications we do not