Distributed Processing Environment Research Articles

Distributed CIM systems — planning for the future

This paper presents some insight into the Computer Integrated Manufacturing (CIM) Systems that operate in a distributed processing environment. It will discuss case studies of distributed CIM implementations currently being undertaken today.

Coordinated Resource Recovery in VM/ESA

A system service for coordinated recovery of resources is a critical function needed for distributed processing environments because applications need to provide for data integrity while the location of the data and processes are transparent to the application. VM is the first IBM operating system to provide Coordinated Resource Recovery as a system service rather than as a service provided by unique environments running on the operating system, and the VM Common Programming Interface-Communications and Shared File System are the first subsystems to utilize the service. This paper is an overview of why and how VM provided Coordinated Resource Recovery (CRR). CRR is the implementation of the Systems Application Architecture™ (SAA™) resource recovery interface within Virtual Machine/ Enterprise Systems Architecture™ (VM/ESA™). This coordinated sync point system service allows one or more applications or subsystems to update multiple resources and to request that all updates be committed or backed out together. The applications and their respective resources can be local or distributed. CRR either coordinates the request to commit or backout immediately, or supports automatic resource resynchronization in case a system or subsystem fails. When restart is not possible, CRR allows for intervention by a system operator or administrator.

On distributed systems performance

The behavior of two interacting processes in a distributed processing environment is analyzed. This problem represents a class of problems which we will confront in the next decade as distributed systems are implemented.

DVLASIC: catastrophic fault yield simulation in a distributed processing environment

Simulation of local process disturbances is a computationally intensive task. The VLASIC (VLSI LAyout Simulation for Integrated Circuits) catastrophic-fault yield simulator uses a Monte Carlo method that often requires tens of CPU hours to perform a simulation. In order to reduce the simulation time, DVLASIC, the distributed-environment version developed by the authors, achieves a speedup of 13.3 over VLASIC, with an efficiency of 89%. The authors describe the distributed processing environment and implementation techniques used to obtain this speedup. The distributed processing environment can also be applied to many other CAD problems. >

A PARALLEL ARCHITECTURE FOR AI NONLINEAR PLANNING

This paper presents a resource-level conflict detection and conflict resolution scheme which is combined with a state-level backward planning algorithm and provides efficient conflict detection and global conflict resolution for nonlinear planning. The scheme keeps track of the usage of individual resources during planning, and constructs a Resource-Usage Flow (RUF) structure (based on which conflict detection and resolution is accomplished). The RUF structure allows the system to perform minimal and nonredundant operations for conflict detection and resolution. Furthermore, resource-level conflict detection and resolution facilitates problem decomposition in terms of resources, thereby providing easy implementation in a parallel and distributed processing environment. Performance analysis indicates that the proposed architecture has a speed-up factor of the average depth of a plan network, D(Na), compared to the distributed NOAH, where Na (the total number of action nodes at the completion of planning) and D(Na) are considerably larger than the number of resources involved in planning as well as the number of initial goal states.

Security of open systems

As an enhancement to ISO's Open Systems Interconnection Reference Model, ECMA defines a security framework for distributed systems which is based on the concepts of security domains and security facilities. This paper describes a functional decomposition of the security of distributed systems, thus providing building blocks which can be specified and analysed in detail and can be used to establish a secure distributed processing environment for both private application software and public communication services such as electronic mail. The aim is to achieve both interoperability (openness) and a high level of confidence.

The Benevolent Bandit Laboratory: a testbed for distributed algorithms

The design, implementation, and use of a distributed processing environment on a network of IBM PCs running DOS is described. Temporarily unused PCs can be accessed by other users on the network to perform distributed computations. An owner of a PC need not be aware that the machine is being used during idle times; the machine is immediately returned when the owner begins to work again. Some degree of computation resiliency is provided in this unreliable environment; if a PC is part of a distributed algorithm and is reclaimed by its owner, the system finds a replacement node (if possible), resends the affected code to the processor, and restarts it. Thus, a distributed computation is able to proceed despite a set of transient processors. System performance, distributed applications, and fault tolerance are discussed. Performance improvements are demonstrated by applications like parallel merge sort and a distributed search solution to the eight puzzle.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Capacity Expansion for Information Flow Distribution in Multi-Path Computer Communication Networks

Abstract:In the distributed processing environment characteristic of the management information systems of today, a need to superimpose an additional information flow onto an existing computer communication network arises frequently. The situation may result from a company’s expanding needs for communication over its private or a shared network, or because a new user company is joining an existing shared network. In order to accommodate the new information flow without negatively affecting the waiting times for the previously contracted communications, network capacity may have to be expanded. We present an algorithmic method for least-cost capacity expansion of a set of network paths preselected to carry the incremental information flows. Owing to the insights gained as a result of extensive computational experiments, we then provide a simplified technique applicable to a broad class of cases. Polynomial time complexity of the algorithms included in the presented methodology points to its practicality.

Multilevel security issues in distributed database management systems

This paper describes the security needs in a distributed processing environment common to many enterprises and discusses the applicability of multilevel secure database management systems in such an environment.

Data modeling in DELAB

As the size and complexity of processing and manufacturing systems increases, the need for Database Management Systems (DBMS) that meet the special needs of studies that experiment with such system...

Alchemy model: An operating system model for a microcomputer-based, heterogeneous processing configuration

In addition to providing raw computer power a distributed processing environment is amenable to incremented design. This inherent design feature means that it is possible to augment existing facilities by ‘simply’ interfacing new resources as and when they become available. Although the revitalisation of existing, ageing, and probably worthless, equipment is very attractive the costs associated with their interconnection are often prohibitive. A design model is presented which aims to minimise these costs by maximising design visibility.

ISDN Architecture

Modern communications-based applications require a degree of flexibility, user control, responsiveness, and integrative capability beyond that provided in traditional networks. Solutions founded on digital connectivity implemented in a distributed processing environment have begun to meet these needs. These solutions have been constrained, however, by limits to the extent in which systems could communicate with each other across network boundaries. The Integrated Services Digital Network architecture defines key interfaces that support and extend this distributed processing capability and do this in an environment of internationally defined interface standards. These interfaces are channelized into two broad categories: information-bearing channels and signaling channels. Protocols for the latter have been defined using the layered architecture of the open systems interconnect reference model. Two specific ISDN access arrangements have been specified and are planned for systems within an overall ISDN architecture. Applications for such an architecture and future directions are discussed.

An Experimental Study of Software Metrics for Real-Time Software

The rising costs of software development and maintenance have naturally aroused intere5t in tools and measures to quantify and analyze software complexity. Many software metrics have been studied widely because of the potential usefulness in predicting the complexity and quality of software. Most of the work reported in this area has been related to nonreal-time software. In this paper we report and discuss the results of an experimental investigation of some important metrics and their relationship for a class of 202 Pascal programs used in a real-time distributed processing environment. While some of our observations confirm independent studies, we have noted significant differences. For instance the correlations between McCabe's control complexity measure and Halstead's metrics are low in comparison to a previous study. Studies of the type reported here are important for understanding the relationship between software metrics.

Design and implementation of a distributed processing environment with personal computers connected to a time-sharing system

As personal computers become more and more popular, it is common to connect individual personal computers to a host computer system to create a distributed processing system. This paper describes the design and implementation of such a data-communication system, PCTCom, based on connecting a microcomputer to a host time-sharing system. PCTCom is modulated into three layers: the physical layer, the data-link layer and the application layer. In the physical layer, the personal computer is connected to the host computer as a terminal via a serial link. The implementation of the other two layers are accomplished by software. PCTCom is simple and can be implemented with little or even no additional hardware cost. Although several difficulties were encountered in our experiments, PCTCom does function correctly.

The ISO connectionless transport standards

The term Open Systems Interconnection (OSI) has increased in familiarity to those involved in distributed processing and data communications to such an extent that an understanding of the architecture it describes and the principles behind its development have quite nearly become commonplace. OSI consists of two closely related pieces: the OSI Basic Reference Model, ISO 7498, which provides a coordinated framework for the development of protocols to support distributed processing environments; and the set of related protocols which achieve that objective. The Reference Model describes an architecture wherein functions which provide communications are separated into a hierarchy of seven layers. Each layer provides services to the next higher layer, utilizing the services provided by subordinate layers in order to achieve that end. At each layer, protocols are employed to convey information between peer layer entities in communicating systems. When in compliance with the suite of standard protocols prescribed for OSI, systems are considered to be "open".

The ISO connectionless transport standards

The term Open Systems Interconnection (OSI) has increased in familiarity to those involved in distributed processing and data communications to such an extent that an understanding of the architecture it describes and the principles behind its development have quite nearly become commonplace. OSI consists of two closely related pieces: the OSI Basic Reference Model, ISO 7498, which provides a coordinated framework for the development of protocols to support distributed processing environments; and the set of related protocols which achieve that objective. The Reference Model describes an architecture wherein functions which provide communications are separated into a hierarchy of seven layers. Each layer provides services to the next higher layer, utilizing the services provided by subordinate layers in order to achieve that end. At each layer, protocols are employed to convey information between peer layer entities in communicating systems. When in compliance with the suite of standard protocols prescribed for OSI, systems are considered to be "open". Initially, ISO 7498 defined only the requirements of connection mode services within the Reference Model. However, in recognition of the importance which connectionless-mode applications play in the data communications environment, an addendum to ISO 7498 was produced which defines the requirements for connectionless-mode services in the Reference Model. The purpose of this paper is to present an overview of the development of connectionless-mode service and protocol within the Transport Layer of the OSI Reference Model.

An analytic orbit prediction program generator

An ephemeris program generator is designed to automatically construct compact efficient ephemeris programs tailored specifically for tasks encountered in online real-time computing environments. For portability the ephemeris programs are written in standard FORTRAN IV. The analytic theory underlying the ephemeris program is the theory developed by Alfriend and Coffey which solves the zonal satellite problem is closed form. The flexibility required by a diverse user community is built into a meta-program which automatically constructs the ephemeris programs according to a program template/The program template is dictated from a terminal according to the specifications of the particular application. Using the program template the meta-program transcribes the series representing the theory into FORTRAN-compilable code. The largest version of the current ephemeris program can be held in the memory of current microcomputers and could be adapted for use in multiprocessing or distributed processing environments.

Experiments in Automatic Microcode Generation

A procedure is described which permits applications problems coded in a higher level language to be compiled to microcode for horizontally microprogrammed processors. An experimental language has been designed which is suitable for expressing computationally oriented problems for such processors in a distributed processing environment. Source programs are compiled first to a machine independent intermediate language and then to a machine dependence form consisting of elementary microoperations, with optimizations performed during each step. The microoperations are then compacted into executable microinstructions for a specific target machine. The procedure has been implemented for experimental purposes and used to compile several different types of applications programs. The experimental results are presented with an interpretation and analysis, along with recommendations for future study.

A layered architecture for a programmable data network

The Advanced Communication Systems AIS Net 1000 service offered by American Bell Inc. provides a service that combines Data Communications with Data Processing capabilities. The Net 1000 network offers customers a nation-wide service that can be shared among multiple customers. It interfaces with a wide variety of terminals and hosts, and allows incompatible devices to communicate with each other. It provides call, store and forward, and storage services. It also provides customer-programmable capabilities in a distributed processing environment. With the terminal and host computer compatibility, customer programmability, and shared public networking capability, this data network offers &lt;u&gt;the&lt;/u&gt; service for the 80's. The presentation will describe the objectives, approach, and capabilities of the AIS Net 1000 service architecture. It will also address the operating environment and the resulting architecture of the call, store and forward, and storage services.

On Optimal Allocation in a Distributed Processing Environment

Distributed processing has been motivated by many objectives. Among these are a desire to share resources, reduce communications costs (as compared to a centralized processing scheme), increase performance and decrease response time by partitioning tasks and achieve higher system availability (fault tolerance). Typically, processes (tasks) within a single processor and across processors will communicate among themselves in such a distributed environment. It is desirable to be able to assign tasks to processors in some optimal manner. In this paper, a quadratic partitioning model is developed to represent this interaction of tasks in a distributed processing environment. Initially, no capacity constraints are considered at any processor. An optimal solution (in a probabilistic sense) is found for this uncapacitated problem using a probabilistic branch and bound technique. Capacity constraints for processors are then introduced. Three intuitively appealing heuristics are developed to obtain good, though not necessarily optimal, solutions to the capacitated problem.