HIPPARCH: Summary HIPPARCH aims at investigating a novel architecture for high performance communication protocols. We envisage the emergence of a wider and wider range of applications (realtime, multicast, mobile are just 3 paradigm shifts that are causing communications software engineers to re-evaluate traditional approaches). At the same time, monolithic software techniques are increasingly frowned on in the industry. Modularization and dynamic integration through new language techniques is becoming commonplace in other application domains. Hence we believe that a new standardization phase is needed wherein we discard the approaches of 1960s software and networking, and use modern formal definition and system structuring techniques, and use specification and implementation languages. The HIPPARCH architecture based on the Application Level Framing (ALF) and Integrated Layer Processing (ILP) concepts developed at MIT alf . In a nutshell, ALF states that applications should organize the data they are sending as a sequence of autonomous ``frames'', which will be at the same time the unit of transmission, the unit of control and the unit of processing. This allows for efficient design of application specific protocols, while minimizing queuing delays. It also enables the integration of multiple transmission control layers in a single loop, maximizing the efficiency of modern processors. This new architecture will be a very good fit for the emerging information infrastructure. We believe that the future Internet will be very heterogeneous,i.e. a mix of conventional networks, high speed fiber optic based circuits and relatively low speed mobile and wireless networks. The application requirements will also be very heterogeneous. We will see a mix of conventional data and real time signals, point to point and point to multipoint transmissions. ALF and ILP will enable us to efficiently design application specific protocols, integrating controls tuned to the applications needs and adaptation loops that take the best of available network resources. Four of the HIPPARCH partners have been cooperating since January 1994 to investigate this architecture, in a modestly funded effort. We obtained encouraging results. We applied the ALF/ILP design rule to a set of representative applications, and observed results which validated the initial hypothesis. We are developing a prototype version of an ALF protocol compiler, based on the ESTEREL language. In HIPPARCH phase 2 we propose to extend these results to the broadest possible range of information services and clients. We will study benefit of the ALF/ILP principles for the design of communication systems supporting multicast, mobility, and real-time media. The main results will be: A protocol compiler to be made available to partners and other research groups. Specific control algorithms for unicast and multicast ``network conscious applications'' mixing real-time and conventional signals. An execution environment which supports real time protocol implementations generated by the compiler. itemize A relevant application consisting of an architecturally enhanced WWW client and server will be developed to serve as the show-case of the project. The application will demonstrate the viability of the compiler and the performance of the adaptive control algorithms. The project results will be used in various environment, including mobile and wireless networks. The project results will be distributed within standardization bodies for discussion and possible standardization. The project includes industrial partners with experience in producing networked information services. Often, the path from research prototype to commercial product is slowed by the need for industry staff to understand the internals of the entire new system, and occasionally even to re-write it. We want to automate this transition, and to do so for a system of significant size. We will extract some general lessons for a protocol environment in which the "design to product" cycle for new services and applications is radically reduced in time and cost.