DeleGate is a general purpose application level gateway which relays a variety of application protocols on the Internet, including HTTP, FTP, Gopher, WAIS, NNTP, SMTP, POP, Telnet, X Window and CU-SeeMe. It overcomes a variety of obstacles on the Internet caused by barriers for security, insufficient band width and mismatching among protocol specifications and implementations. An application level gateway mediates communication between clients and servers recognizing the syntax and semantics of application protocols it is relaying. Therefore it can supply various functions like caching and translation to an application protocol, for its procedure and data, without changing clients and servers. This paper presents an overview of DeleGate and its two unique features termed mounting and connection cache. Mounting is a mechanism which provides a virtual view for a name space of networked resources by filtering, aliasing and merging names from original name spaces. Resource names transferred as data on each protocol are recognized and rewritten by DeleGate. Connection cache allows established connections to servers to be resources which are sharable and reusable among clients. It reduces the cost of repetitive connection establishments to the same server, where the repetition is introduced by an application protocol gateway which relays between different protocols.
Object-stacking is a model for structuring object-based systems. The main feature of the object-stacking model is that layers of objects with a uniform interface are constructed, and the functions of these objects are integrated. In object-stacking, filtering functions are converted into objects. By implementing object-stacking in the World Wide Web, several filtering functions scattered all over the world can be integrated into a single web resource. In object-stacking on the World Wide Web, the superfluous recursive stacking problem should be solved. This paper shows the solution of the problem by using redirection HTTP objects.
Current research into architectures for delivery of multimedia to the home have concentrated on systems where a user requests an object and it is delivered in real time: an "on-demand" or "client pull" system.This paper describes a new architecture that delivers multimedia objects asynchronously when the description of the object matches their interests and preferences: a "server push" system.The push-media approach to delivery of information is gaining popularity. Examples of the use of this approach will be discussed. Our push-media architecture will then be described including several example systems: a music-on-order system, a personalised audio news system and a listener constructed radio programme.
VRML is a language to describe virtual 31) spaces. In order to achieve reuse and sharing of data described in VRML among multiple applications, we designed and implemented a VRML-object database system. In our system, we can add semantic information onto VRML data, and can issue queries by using that information. We also introduce the concepts of spatial view and level-of-detail based access control. When a user issues a query, the system produces a spatial view by reorganizing a virtual space in accordance with the user's concern and the user's access capability.
Intelligent agent systems are developing into an important paradigm in Artificial Intelligence and in general computer science. An agent can be seen as an autonomous system that perceives and acts on its environment, performing tasks and pursuing goals independently of user control. Recently, there has been much research interest directed toward the design of multi-agent systems that act collaboratively on tasks that are too complex for any single agent to perform on its own. We discuss the design and implementation of teamwork amongst autonomous agent systems. In particular, we describe an approach, using team plans, that addresses a number of important issues, such as: balancing autonomous behaviour with commitment to team behaviour; team formation; distribution of tasks amongst team members; coordination and synchronisation of actions.
When writing a program to be executed in the worldwide object-oriented distributed systems environment, the source code may include class names developed at other sites. In this case, using the class name as a runtime class identifier, which should be globally unique, may cause problems such as: 1) The same class name representing a different class is found in another site. 2) Classes using different versions of a class cannot work together. To tackle these problems, the authors proposed and implemented a local class name facility called ‘chool. The school facility separates the compile time class name and the runtime class identity. This paper describes our design and implementation of the school and the runtime class identity in the OZ++ system: a worldwide object-oriented distributed systems environment developed by the authors.
Computer support for meeting scheduling is a focus of active research in several areas, including distributed AI and intelligent agents, software engineering, and information systems. It is also of very high applied interest: After a decade of very modest success, commercial scheduling software is taking hold in some organizational settings. However, meeting scheduling is most significant for being one of the first desktop group support applications that is widely used in some large organizations. The slow acceptance of most groupware applications raised questions about the conditions under which successful use would develop. Studies of calendar use and meeting scheduling at Sun, Microsoft, and other organizations provide clear answers to the conditions under which such use can develop. In addition, differences in the patterns of use within and across organizations provide further insight into the relationship between a widely-used technology and the behaviors that accompany it. Relatively minor choices in the way groupware is rolled out could have substantial impact on the way it is used.
Mobile agents are programs that can be dispatched from one computer and delivered to a remote computer for execution. Arriving at the remote computer, they present their credentials and obtain access to local services and data. They also provide a single uniform paradigm for distributed object computing, encompassing synchrony and asynchrony, message-passing and object-passing, and stationary objects and mobile objects. In this paper, we describe our Java-based mobile agents called Aglets and present its programming interface, called Java Aglet API.
Service proxy framework is a framework for constructing adaptive mobile applications based on the client-mediator-server architecture. The goal of our framework is to provide an infrastructure for building adaptive mobile applications. Using the service proxy framework, an application can change its behavior when the operational environment is changed dynamically. This paper presents host mobility supports for the service proxy framework. In our approach, a mobile application can be dynamically adapted to the characteristics of underlying networks when the handoff of a mobile host occurs. We especially focus on a robust handoff protocol for mobile applications using the service proxy framework. This project is funded by the Advanced Information Technology Program (AITP) of Information-technology Promotion Agency (IPA), Japan.
In this paper, we describe a dynamic extensible kernel architecture suitable for Real-Time Mach kernels. We design a new kernel architecture, the micro-kernel tray architecture. Our goal is to provide a new kernel foundation that enables an application to extend kernels dynamically and also enables an operating system builder to create a micro-kernel construction kit similar to a compiler construction kit. A micro-kernel tray provides a software backplane bus for micro-kernel based systems. A kernel construction kit would save even more time and would also enable the developer to focus her/his effort on the interesting aspects of operating systems. We report the preliminary step of our research, propose dynamic loadable object support for Real-Time Mach kernels, and discuss the issue of dynamic reconfigurable kernels.
MESSENGERS is a system for general-purpose distributed computing based on the concept of autonomous mobile objects, capable of navigating through a network and invoking native-mode C functions in the nodes they visit. We present performance measurements using three specific applications to illustrate the cost of using this paradigm for developing and using distributed applications.
Today almost all database systems use B-trees as their main access method. One of the main drawbacks of the classical B-tree is, however, that it works well only for one-dimensional data.In this paper we present a new access structure, called UB-tree (for universal B-tree) for multidimensional data. The UB-tree is balanced and has all the guaranteed performance characteristics of B-trees, i.e., it requires linear space for storage and logarithmic time for the basic operations of INSERT FIND DELETE. In addition the UB-tree has the fundamental property, that it preserves clustering of objects w.r., to Cartesian distance. Therefore, the UB-tree shows its main strengths for multidimensional data. It has very high potential for parallel processing. With the new method, a single UB-tree can replace an arbitrary number of secondary indexes. For updates this means that only one UB-tree must be managed instead of several secondary indexes. This reduces runtime and storage requirements substantially. For queries and in particular range queries the UB-tree has multiplicative complexity instead of the additive complexity of multiple secondary indexes. This results in dramatic performance improvements over secondary indexes.The UB-tree is obviously useful for geometric databases, datawarehousing and datamining applications, but even more for databases in general, where multiple secondary indexes are widespread, which can all be replaced by a single UB-tree index.
Rapid growth in networks and related services have resulted in demands for more effective, accessible, intelligent and scalable network management systems. Not only managers but also users need access to network management information. Traditional network management architectures are utterly inadequate to meet these requirements. In this work we describe VINES - a next generation network management system that builds on state-of-the-art technolgy and latest results. VINES uses a new web-based management architecture, that is intelligent and distributed. A set of Management Domain Servers, implementing an elaborate access control mechanism, allows access to management information without compromising security. To realize the new architecture a new knowledge base framework is designed, highly effective detection and diagnosis algorithms are employed, and the resilience of the management system to network failures is ensured. The results of an experimental prototype are discussed.
Despite nearly 20 years of progress toward ubiquitous computer connectivity, distributed computing systems have only recently emerged to play a serious role in industry and society. Perhaps this explains why so few distributed systems are reliable in the sense of tolerating failures automatically, guaranteeing properties such as performance or response time, or offering security against intentional threats. In many ways the engineering discipline of reliable distributed computing is still in its infancy.
HORB is a Java ORB (Object Request Broker) that extends Java for distributed object oriented computing. Applications of HORB and HORB itself are network portable and transferable, since HORB provides both portability and interoperability among different kinds of OSs. This paper describes the organization and the main features of HORB. It provides remote object creation/connection, remote method call, object passing, security, and so on. The usage of HORB is easy and simple. Performance evaluation shows that HORB is two or three times faster than javaRMI.
Recently, we can find three remarkable features in advanced computer networks. First is the development of client software systems based on the World-Wide Web (WWW). Second is the spatial extension of communication environments through wireless communication technologies, making mobile computing possible. Third is drastic increase of available bandwidth by high speed network technologies such as ATM (Asynchronous Transfer Mode) switching. In this paper, to facilitate the development of databases based on these three network software/hardware technologies, we discuss their architectural issues. They are the issue of three-tier database architecture based on WWW clients and network integration servers, the issue of database architecture for a mobile computing environment, and the issue of database migration techniques for ATM based networks.
The desktop conferencing system described here is called WISH (for Web Information SHare) and enables users to conduct a conference on the Internet. Users can participate in the conference by using WWW browsers and can share HTML documents with all participants in realtime. When one participant accesses WWW pages and browses them, the same pages are automatically displayed on all other participants' browsers. Participants can also annotate directly on the pages and share the annotations in realtime.
We briefly summarize the mobile computing environment from a computer science point of view, and discuss some of the interesting possibilities raised by the prospect of extremely large numbers of computational nodes networked over wireless channels. Among these are: the use of reflection and the metaspace construct to deal with some of the Quality of Service and Connectivity problems of mobile computing, the problem of dealing with the degraded user interfaces often implied by mobility, and the possible applicability of mobile objects to mobile computing.
With the increasing interest in multimedia, researchers across various disciplines, in particular, the database community and image processing community, have teamed up to conduct research in building multimedia database systems through integration of available technologies in various areas. In this paper, we introduce a hybrid approach to multimedia database systems in which traditional database systems are augmented by added media-dependent modules to manipulate and search multimedia data. SEMCOG (SEMantics and COGnition-based image retrieval) is a multimedia database system based on this hybrid architecture. We use it as an example to illustrate how to design and implement a hybrid multimedia database system. We also give our experiences with prototyping SEMCOG and issues raised for future research.
