State-space caching is a verification technique for finite-state concurrent systems. It performs an exhaustive exploration of the state-space of the system being checked while storing only all states of just one execution sequence plus as many other previously visited states as available memory allows. So far, this technique has been of little practical significance: it allows one to reduce memory usage by only two to three times, before an unacceptable blow-up of the run-time overhead sets in. The explosion of the run-time requirements is due to redundant multiple explorations of unstored parts of the state-space. Indeed, almost all states in the state-space of concurrent systems are typically reached several times during the search. In this paper, we present a method to tackle the main cause of this prohibitive state matching: the exploration of all possible interleavings of concurrent executions of the system, which all lead to the same state. Then, we show that, in many ...

Multipath routing and data retransmission cause data misordering. Although data could be reordered before processing, it is simpler and more efficient to process data as it arrives. Because existing protocol functions generally cannot process misordered data, we need new functions with minimal ordering constraints. For two example functions, CRC error detection and CBC mode encryption, we show that the functions have ordering constraints and we present new functions that provide similar functionality without ordering constraints. Keyword Codes: C.2.2 Keywords: Network Protocols 1 Introduction While developing a high-speed protocol processing host-network interface for the AURORA gigabit network, we came across an interesting problem associated with packet misordering. Some protocol functions cannot be computed on misordered data, and thus, the data must be reordered before processing. We demonstrate the problem using cyclic redundancy code (CRC) error detection and cipher block chain...

As advances in technology enable us to implement very high speed computer networks, we expect to use our networks for more diverse applications. While the Internet was designed with textual data processing in mind, future networks will carry information such as voice, music, images, and video, along with textual data. Many new applications will have real-time performance requirements, where the timing of data arrival is crucial to its usefulness. This paper describes a methodology developed at the University of California at Berkeley to support such applications, reviews related research work, and proposes a real-time delivery system, composed of a new protocol for administration of real-time connections, combined with modifications to the Internet Protocol (IP) to support such connections. Transport protocol requirements are also discussed. This work is intended to facilitate experiments with real-time communication over the Experimental University Network (XUNET). 1. Introduction ...

When synthesizing control-flow dominated descriptions based on VHDL, different styles of semantically equivalent descriptions may differ significantly in quality. This paper discusses the effect of the input description on High-Level Synthesis when using VHDL. In order to show this effect, a high speed protocol based on the ISO reference protocol Abracadabra is used. Five VHDL descriptions styles of the same protocol have been synthesized using AMICAL, a VHDL based behavioral synthesis tool. Discussions of the different results leads to a VHDL based methodology for protocol modelling in order to produce efficient designs.

x-ATM is a portable toolkit of Asynchronous Transfer Mode (ATM) protocols implemented in a general purpose communications kernel (the x-kernel). The toolkit provides an environment for experimentation with ATM adaptation layers, interoperable ATM signalling, IP over ATM, and real time transport protocols. The x-kernel permits easy composability and promotes protocol modularity, thus x-ATM protocol modules may be arbitrarily layered on top of anchor protocols that represent different hardware devices. This supports x-ATM's protocol suite portability across many different platforms. The Unix[21] version of the toolkit currently has hardware support for Fore Systems SBA-100 host adapters, ASX-100 ATM switch and Fore ATM API, as well as the iPoint[16] experimental ATM switch, with support for other hardware platforms planned. The toolkit may be used to implement ATM signalling entities used in ATM switch controllers and hosts with ATM adapters. In our experience, x-ATM has proven to be a ...

A traditional protocol implementation typically consists of at least two distinct parts, a sender and a receiver. Each part runs on a distinct machine, with the implementation provided by a local expert. At best, the two machines are of the same type and the protocol implementations are provided by the same person. More likely, however, the machines are not of the same type and the implementations of the two halves of the protocol are provided by two different people, working from an often loosely defined protocol specification. It seems almost unavoidable that the two implementations are not quite compatible. In this paper we consider an alternative technique. With this method, one of the two implementors can design, formally validate, and implement all the relevant protocol parts, including those parts that are to be executed remotely. Each communication channel is now terminated on the receiving side, by a single standard protocol interface, which can be called a Universal Asynchronous Protocol Interface or UAPI. Though it is likely that the UAPI is most efficiently implemented in hardware, it can also trivially run as a software module, e.g. under a standard UNIX ® operating system (in our case under 10th Edition Research Unix). This paper introduces the concept of a UAPI and explains how the sample software controller was constructed.

recently as a new B-ISDN ATM Adaptation Layer (AAL) protocol standard, initially for use in the Signaling ATM Adaptation Layer (SAAL), but also for support of certain types of user data transfer. SSCOP is a new type of protocol, embodying several design principles for high speed link and transport layer protocols. In this paper, the basic operation of SSCOP is described and the SSCOP design is compared with other similar protocols. Next, the relationships between key protocol parameters (protocol window, control message transmission interval) and maximum achievable throughput efficiency are explored. In particular, approximate performance equations are derived for predicting the maximum throughput efficiency of SSCOP based on the selected environment and parameter settings. The equations can be used to determine how much buffer capacity and/or what protocol timer settings allow SSCOP to operate at high performance. The analytical results are confirmed through comparison with simulation results. In addition, simulation results illustrate the high throughput performance achievable when using SSCOP in a highly errored or lossy environment. I.

this paper confirm the results of [6] which illustrate that sufficient receiver buffer size is one of the key requirements for efficient operation of a selective retransmission protocol. When a satisfactory resequencing buffer is available, the protocol can offer adequate performance even in very degraded environments

State-space caching is a verification technique for finite-state concurrent systems. It performs an exhaustive exploration of the state space of the system being checked while storing only all states of just one execution sequence plus as many other previously visited states as available memory allows. So far, this technique has been of little practical significance: it allows one to reduce memory usage byonlytwo to three times, before an unacceptable blow-up of the run-time overhead sets in. The explosion of the run-time requirements is due to redundantmultiple explorations of unstored parts of the state space. Indeed, almost all states in the state space of concurrent systems are typically reached several times during the search.

this paper we consider an alternative technique. With this method, one of the two implementors can design, formally validate, and implement all the relevant protocol parts, including those parts that are to be executed remotely. Each communication channel is now terminated on the receiving side, by a single standard protocol interface, which can be called a universal asynchronous protocol interface, or UAPI