A distributed graphics library system

We present a set of library routines that allow easily parallelized graphics rendering routines that require no communication between each parallel task, such as ray-tracing, to be run efficiently in an environment of distributed workstations. The presentation of the paper focuses on the problems encountered in implementing a distributed system under Unix and proposes solutions to each problem. Specifically, we discuss the challenges involved in overcoming the limits of communicating with a large number of processes in Unix and in providing fault tolerance when using sockets. Technical aspects of the implementation and some additional problems that were encountered are discussed. Finally, we compare the rendering times for a complex image with a renderer using the library and show that the library routines are able to exploit much of the existing parallelism. The library is presented using a graphics application, though the concepts are generic enough to be of use in designing any distributed system under Unix.     

Banet — a local area network for distributed data processing

A broadcast architecture network (Banet) suitable for distributed data processing is proposed. One feature of Banet is that the broadcast-within-a-group function is supported not only by the datalink level but by the transport or session-level network structure. The commitment control scheme is included in the network protocol. Design goals, physical structure and protocols are discussed.     

Designing a distributed database on a local area network: a methodology and decision support system

Local area networks (LANs) are important for an enterprise to hold a competitive edge. Many companies have therefore converted terminal-based computing systems to LAN-based distributed data processing systems. This paper proposes a design methodology for distributed databases connected by a LAN. Two primary objectives of the methodology are: (i) to allocate data files and workload among heterogeneous servers; and (ii) to determine the number of servers to satisfy the response time required for processing each transaction. The file and workload allocation decision is formulated as a nonlinear zero–one integer programming problem. This problem is proven to be NP-complete. A heuristic is developed to solve this problem effectively. A decision support system is implemented and an example is solved to illustrate the practical usefulness of the system.     

A local area network architecture for communication in distributed real-time systems

Distributed real-time systems of the future will require specialized network architectures that incorporate new classes of services and protocols in order to support time-constrained communication. In this paper, we propose a new local area network architecture for such systems. This four-layered architecture is characterized by new classes of connection-oriented and connectionless services that take into account the timing constraints of messages. We describe various aspects of the logical link control layer of the architecture and various real-time protocols that may be employed at the medium access control layer in order to support the new classes of services. We also describe a homogeneous approach to the implementation of medium access control protocols to support both connection-oriented and connectionless services, based on a uniform window splitting paradigm.     

A communication system supporting large datagrams on a local area network

This paper describes an implementation of a software interface designed for use from within UNIX

1 UNIX is a trademark of A T & T Bell Laboratories.

application programs for network communications. This interface provides the abstraction of possibly very large datagrams, supports ‘scatter-gather’ facilities, and maintains standard network addresses consisting of (host number; port number) pairs. This interface has been provided on different data communication facilities, allowing uniform program access to those facilities. The particular implementation developed for a local area network is described, and performance results obtained are examined and compared with those obtained from a conventional datagram interface to that network. The results obtained confirm that the abstraction of very large datagrams enables the construction of an efficient mechanism for process-to-process communications over the network that is not only more convenient, but also has significant performance advantages over the use of conventional (small) datagrams.     

Choosing a local area network

As they employ many different technologies, it is difficult to give a precise definition of a local area network. Most LANs, however, have certain characteristics and these are discussed as are the reasons for using a LAN. Although the technology employed by LANs is not as important to the user as the level of service offered, the technology does have some bearing on this service. The two major types of LANs currently being marketed are described. The applications of LANs and considerations that should be borne in mind when choosing a LAN are discussed and the potential pitfalls pointed out.     

Using a local area network

Local area networks (LANs) are high speed multiple access networks which have been developed to meet the needs of distributed computing and the use of electronic equipment in the office. They have grown in use as the use of the microprocessor has proliferated. There are two basic types of LAN; baseband and broadband. Ethernet is one of the former types. LANs have five main areas of application; distributed computing, office systems, terminal networks, factory systems and micronetworks.     

An interactive network graphics system

This paper discusses a number of issues relevant to the design of systems for performing interactive graphics over a computer network such as the ARPA Network. We then describe a working implementation of such a system and an example of its use.     

基于本地网的分布式蠕虫检测系统设计

为能有效检测本地网中的已知蠕虫和未知蠕虫,设计了一个分布式蠕虫检测系统.探讨了系统的部署策略和结构,并详细描述了系统检测算法的设计过程.检测算法分为可疑主机检测和感染主机检测两个阶段,前者通过监控主机的网络连接异常发现可疑主机,后者采用误用检测和关联分析判断可疑主机是否为感染主机.仿真实验结果表明了该系统的有效性.     

DGLa: A Distributed Graphics Language

A distributed graphics programming language called DGLa is presented,which facilitates the development of distributed graphics application.Facilities for distributed programming and graphics support are included in it,It not only supports synchronous and asynchronous communication but also provides programmer with multiple control mechanism for process communication.The graphics support of DGLa is powerful,for both sequential graphics library and parallel graphics library are provided.The design consideration and implementation experience are discussed in detail in this paper.Application examples are also given.     

Token-ring local area network

A local area network (LAN) should be easily accessible, extremely reliable, and extendable in both function and physical size. The star/ring wiring topology with token-access control has emerged as a technology that can meet all of these objectives. The requirements of small networks with just a few terminals, as well as those of very large networks with thousands of terminals, can be achieved through this one architecture. The paper is a summary of the fundamental aspects of the architecture, physical components, and operation of a token-ring LAN.     

Development of a low-cost local area network

This article describes the development of a LAN in the University of Hull's Department of Psychology. Financial constraints led to the need to build the LAN around inexpensive components. The Clearway system was chosen for its low cost and flexibility and has provided a good solution to the particular combination of communications problems which had to be solved. The article describes the way the design of the system was related to the needs and resources of the Department and how a software ‘front-end’ was developed to make the system accessible to a wide range of users.     

Myrinet: a gigabit-per-second local area network

The Myrinet local area network employs the same technology used for packet communication and switching within massively parallel processors. In realizing this distributed MPP network, we developed specialized communication channels, cut-through switches, host interfaces, and software. To our knowledge, Myrinet demonstrates the highest performance per unit cost of any current LAN     

Platon: a university local area network

Platon is a local area network set up in the Department of Electrical Engineering, University of Ottawa, operating at 1 Mbit/s and interconnecting departmental computers, terminals, peripherals, various instruments and the university mainframe. The primary objective of Platon is to facilitate network evolution and provide resource sharing by supporting various application protocols such as database access and file transfer. In addition, Platon is intended to serve as a research vehicle for distributed systems studies. To this end, a packet-voice system and a packet-radio gateway are concurrently being developed. Platon has adopted the bus architecture, using a coaxial cable with CSMA/CD as its access protocol. Retransmissions of collided packets are scheduled using a linear incremental backoff (LIB) algorithm.     

A distributed semantic network model for a collaborative intelligent system

A methodology based on topology theory to model a semantic network for a collaborative system is given. This framework is used to support the creation of a semantic network and to define the associated intelligent cooperative system. Our methodology is illustrated via a set of agents whose knowledge-base is a semantic network. By a series of functions applied on a base of entities, issued from the application domain, a family of sets are synthesized with their subspaces correlated. The resultant subspaces and their relations form a network of elementary and complex concepts that can be naturally represented with the IDEF1x language. A prototype Multi-Agent System (MAS), set up with the Zeus platform,¹ was developed for the Process Plan domain, which was used as a case study. Full correspondence among the subspaces, the semantic network IDEF1x information model and the MAS implementation is obtained by employing this framework.     

A queueing network model for a distributed database testbed system

A queuing network model for analyzing the performance of a distributed database testbed system with a transaction workload is developed. The model includes the effects of the concurrency control protocol (two-phase locking with distributed deadlock detection), the transaction recovery protocol (write-ahead logging of before-images), and the commit protocol (centralized two-phase commit) used in the testbed system. The queuing model differs from previous analytical models in three major aspects. First, it is a model for a distributed transaction processing system. Second, it is more general and integrated than previous analytical models. Finally, it reflects a functioning distributed database testbed system and is validated against performance measurements     

The EENET: A broadband local area network

The EENET is a local area computer network designed explicitly for real-time applications. The formal specifications of the EENET are presented with the analysis of a simulation model for the protocols used.     

A communication system for distributed computations on a network of personal computers

This paper deals with a software system PCCOM which makes it possible to simulate a distributed memory environment for parallel computations by using a local area network of personal computers. This system consists of FORTRAN subroutines that can be used from application programs. Parallel computations may be performed on a network of personal computers under DOS or any other operating system. This software package simulates, partially, the widespread PVM (Parallel Virtual Machine) package which runs under UNIX. A simple example of an application is given here.     

Simple token ring local area network

The implementation of a simple token ring system is described. A token ring is a form of local area network in which hosts must capture the token, a unique bit pattern, to use the communications medium. Many existing token ring implementations designate a separate host as a monitor station for the ring. Thus ring initialization and recovery are centralized, so it the monitor is down the ring becomes unusable.Our aim was to build a simple cheap token ring which would have advantages over an Ethernet-style local area network in terms of guaranteed access time upper bounds. We also attempted to decentralize initialization and recovery by giving each host the ability to initialize and repair the ring.