James Peyton Gray

SNA Networks of Small Systems

This paper discusses SNA/LEN, a possible extension of Systems Network Architecture intended to allow peer, dynamic, and easy to use networking functions for a variety of node sizes down to and including the new generation of personal computers. After reviewing the special requirements posed by small systems and the need for including them as equal partners in networks, the LEN architecture is described. This is done by following a sequence of steps that begin when a logical resource at some node that may not yet be part of the connectivity of an existing network requests a session with a remote logical resource of unknown location. After connectivity with the preexisiting network is established, directory services locates the remote object, route selection services determines the preferred path, a session is activated, and deadlock-free flow control assures a useful flow of data. Preliminary quantitative results from an LEN prototype are presented.

ATM cell encryption and key update synchronization

This paper presents a data compaction/randomization based approach as a mode of block encryption for ATM (Asynchronous Transfer Mode) cells. The presented approach converts a plaintext into pseudo‐random plaintext before ciphering to conceal patterns in the plaintext. The underlying idea behind this scheme is the Shannons principles of “confusion” and “diffusion” which involve breaking dependencies and introducing as much randomness as possible into the ciphertext. In this scheme, confusion and diffusion are introduced into the system by first compressing the ATM cell payload and then spreading a continuously changing random data over the entire content of the cell. As a mode of operation for block ciphering, this scheme offers the following attractive features:(i) plaintext patterns are pseudo‐randomized and chained with ciphertext (thereby, preventing against “dictionary”, “known plaintext”, and “statistical analysis” attacks), (ii) it is self‐synchronizing, (iii) cell loss has no additional negative effect, (iv) no IV (Initialization Vector) storage is required, (v) it is encryption‐algorithm independent, (vi) there is no cell‐to‐cell dependency (no feedback from previous cells), and (vii) it is highly scalable (i.e., cells from the same stream can be ciphered and deciphered in parallel). This paper also presents a secure mechanism for in‐band synchronization of encryption/decryption key updates using a “marker‐cell” that is carried within the data channel. An important aspect of both the above mechanisms is that they do not require any changes to the ATM cell header or ATM infrastructure.

IBM's Systems Network Architecture

A network is more than simply a collection of machines and communication lines. A properly designed network serves a particular purpose for a particular user or class of users. In order to design networks for different users and purposes while minimizing design effort, one can employ a network architecture. In this chapter we look at the services provided by a network, examine what an architecture is, and then look at Systems Network Architecture (SNA),* its design principles and how it enables a network to provide the required services. Our discussion of the architecture appears in two major sections, transporting data and distributed programming. We have attempted to minimize the amount of jargon in this description of SNA. Necessary new terms are introduced in italics. After looking at the architecture, we conclude with a discussion of how SNA has applied the underlying principles.

SNA's design for networking

Systems Network Architecture (SNA) provides users with services needed in a networking environment: sharing of communication lines, directory services, connection establishment, improved reliability, management services, and an appropriate interface and set of facilities for distributed programming. It provides these services within the scope of a layered architecture that has evolved to the rich, set-functions and products that it supports. The major concepts of SNA are reviewed. The three major functions that a network provides-transporting data, enabling distributed programming, and application-level services-are discussed.<<ETX>>

A Preview of APPN High Performance Routing

After a brief review of APPN—Advanced Peer-to-Peer Networking—and a survey of existing routing techniques, a new SNA approach to routing called HPR—APPN High Performance Routing—is introduced. Topics covered in this overview include HPR function placement within the OSI layered model, priority scheduling for multilink transmission groups, Automatic Network Routing, Rapid Transport Protocol, Adaptive Rate-Based congestion control, the relationship of effective congestion control algorithms to throughput and response time, and HPR’s selection of frame relay as a preferred data link control.