This paper studies the issues of flexibility and scalability in the context of network security. In particular, it concentrates on authentication and key distribution services suited for a variety of communication paradigms, network environments, and end-devices. We present the design criteria, specification, and step-by-step construction of authentication and key distribution services based on experience in the KryptoKnight project. The central goal of the KryptoKnight project was the construction of basic network security functions in a minimal, flexible (thus, versatile) and scalable manner. Protocol minimality (in terms of resource usage) and flexibility are not merely theoretical goals; they have clear advantages in environments where computational resources are limited and connectivity is restricted. KryptoKnight was aimed at such environments: small and anemic wireless devices, simple network and data-link entities, embedded micro-devices and other special-purpose communication equipment and configurations. Furthermore, scalability of protocols makes their deployment possible in the presence of rapid network growth and inter-domain communication.

This paper discusses issues and idiosyncrasies associated with changing passwords and keys in distributed computer systems. Current approaches are often complicated and fail to provide the desired level of security and fault tolerance. A novel and ve

An essential function for achieving security in computer networks is reliable authentication of communicating parties and network components. Such authentication typically relies on exchanges of cryptographic messages between the involved parties, which in turn implies that these parties be able to acquire shared secret keys or certified public keys. Provision of authentication and key distribution functions in the primitive and resource-constrained environments of low-function networking mechanisms, portable, or wireless devices presents challenges in terms of resource usage, system management, ease of use, efficiency, and flexibility that are beyond the capabilities of previous designs such as Kerberos or X.509. This paper presents a family of light-weight authentication and key distribution protocols suitable for use in the low layers of network architectures. All the protocols are built around a common two-way authentication protocol. The paper argues that key distribution may require substantially different approaches in different network environments and shows that the proposed family of protocols offers a flexible palette of compatible solutions addressing many different networking scenarios. The mechanisms are minimal in cryptographic processing and message size, yet they are strong enough to meet the needs of secure key distribution for network entity authentication. The protocols presented have been implemented as part of comprehensive security subsystem prototype called KryptoKnight.>

Most existing designs for two-way cryptographic authentication protocols suffer from one or more limitations. Among other things, they require synchronization of local clocks, they are subject to export restrictions because of the way they use cryptographic functions, and they are not amenable to use in lower layers of network protocols because of the size and complexity of messages they use. Designing suitable cryptographic protocols that cater to large and dynamic network communities but do not suffer from these problems presents substantial problems. It is shown how a few simple protocols, including one proposed by ISO, can easily be broken, and properties that authentication protocols should exhibit are derived. A methodology for systematically building and testing the security of a family of cryptographic two-way authentication protocols that are as simple as possible yet resistant to a wide class of attacks, efficient, easy to implement and use, and amenable to many different networking environments is described. Examples of protocols of that family that presents various advantages in specific distributed system scenarios are discussed.>

A review is presented of the different aspects of heterogeneity that need to be resolved to achieve global interoperability, and specifically, to provide global transport service. The authors concentrate on the problems of interconnecting computer networks and systems that use different protocols, either different subsets of OSI (open systems interconnection) standards or OSI and non-OSI protocols. The issues of protocol conversion are discussed, and integrated naming, addressing, and directory mechanisms for interconnecting OSI and non-OSI domains are outlined. More subtle aspects of interoperability are examined for the case of the OSI transport protocol class 4 in mixed LAN/WAN (local-area-network/wide-area network) environments.>

The internetworking capability of a network is typically provided transparently by protocols below the network application interface. However, network designs have been implemented in which the internetwork function is limited or absent. This paper describes one way to add transparent internetworking to such systems without affecting the existing network hardware and software. The solution consists of using application-level software and gateways that project a domain structure onto application names in such a way that neither the applications nor the underlying protocols need be aware of internetworking. A research prototype of this internetworking scheme has been implemented in the context of the NETBIOS application interface; the paper discusses its key features and the lessons learned.

The impact of various implementations of an end-to-end transport service on the performance of high-volume data transfers in a local-area network is investigated. The impact of using a connection-oriented logical-link control (LLC) protocol below a connection-oriented transport protocol is of particular interest. The performance measure is the throughput which can be achieved between two stations either from memory to memory or from disk to disk. The throughput was determined by simulation for various combinations of differing LLC protocols, disk models and processor speeds. The parameters for the simulation model were determined partially by measurements of an implemented file server. The results show that the performance penalty of a connection-oriented LLC protocol can be substantial, but can be significantly lowered when an acknowledgment accumulation strategy is applied.

This paper describes the design and implementation of filing and printing services in a distributed system based on a token-ring local-area network. The main emphasis is put on the communication aspects of the client/server scenario: roles of a client and a server in a communication protocol, and the integration of communication protocols with applications.

As part of an effort to engineer a security kernel for Multics, the dynamic linker has been removed from the domain of the security kernel. The resulting implementation of the dynamic linking function requires minimal security kernel support and is consistent with the principle of least privilege. In the course of the project, the dynamic linker was found to implement not only a linking function, but also an environment initialization function for executing procedures. This report presents an analysis of dynamic linking and environment initialization in a multi-domain process, isolating three sets of functions requiring different sets of access privileges. A design based on this decomposition of the dynamic linking and environment initialization functions is presented.