Alan E. Baratz

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.

Reliable link initialization procedures

It is known that HDLC (high-level data link control) and other bit-oriented DLC procedures ensure data transmission reliability across noisy transmission media, provided that all frame errors are detected and the link processes are synchronized at initialization. It is shown that the HDLC initialization procedure does not ensure synchronization and allows inadvertent loss of data. Several link initialization procedures are proposed and it is proved that they do ensure synchronization. >

Establishing virtual circuits in large computer networks

Abstract Network routing algorithms generally attempt to provide communication between two nodes by sending data messages along the best or shortest path between them. Unfortunately, in large networks it is difficult to maintain knowledge of such path due to the cost in storage, computation, and communication bandwidth. In an attempt to solve this problem the technique of clustering has been proposed. Clustering generally reduces the cost of routing of by sacrificing optimality. Kamoun has shown that this sacrifice is asymptotically negligible under certain strong assumptions. In this paper we propose a new clustering technique which permits us to obtain optimal paths. However, determining these paths requires some effort and thus the methodology is appropriate only if the paths are then used for many messages in virtual circuit fashion. An application to planar networks gives a quantitative demonstration of obtaining optimal paths with reduced path determination cost.

Subtle design issues in the implementation of distributed, dynamic routing algorithms

Abstract The design of distributed, dynamic routing procedures is highly complex, due to the fact that it relies on cooperation among a number of independent processors located at the nodes of communication networks and the fact that a network may face arbitrary topological changes. As a result, many simple, intuitive, concepts that seem, at first glance, that they should result in correct algorithms do not work. This paper evaluates this theme by exploring a number of examples of this phenomenon. The conclusion is that one must be careful both in the overall design of a distributed algorithm, and in its detailed implementation. This also illustrates the importance of careful formal validation of such protocols, rather than informal, intuitive arguments.

Implementing System/36 advanced peer-to-peer networking

System/36 Advanced Peer-to-Peer Networking (APPN) provides highly dynamic, fully distributed peer networking for low-end processors. It is built upon existing SNA Logical Unit 6.2 and Node type 2.1 support. APPN presents System/36 users with a simplified model of communications. The structure of the APPN subsystem is outlined, with particular emphasis on the integration of APPN functions with existing SNA support. The authors describe how particular aspects of the APPN design have been tuned to the System/36 operating environment.

Fault tolerant queries in computer networks

In computer networks with decentralized control, it is necessary to provide a distributed query mechanism, whereby a node can dynamically discover the location of a remote resource. The authors propose a query mechanism based on flooding with feedback, which uses no more than two messages over any link. They examine the fault tolerant aspects of this distributed query mechanism, operating in a network where links and nodes may fail and show that, under certain conditions, the resource is indeed found. They then construct enhancements to this mechanism, where the conditions under which the resource is found are substantially relaxed, at the cost of some increase in message complexity. >

A Perspective on Shared-memory and Message-memory Architectures

Parallel processing is becoming widely accepted as a key technology in the development of extremely high performance computing systems. Although a wide variety of parallel processor architectures have been proposed over the last five to ten years, to date there is very little quantitative analysis comparing the various alternatives. The proposed architectures essentially fall into two broad categories: message-passing architectures and shared-memory architectures. In this paper we will briefly review the key characteristics of message-passing and shared-memory architectures. We then propose an approach to obtaining a quantitative comparison of these architectures and suggest areas for research.