This paper presents the design and implementation of the Intentional Naming System (INS), a resource discovery and service location system for dynamic and mobile networks of devices and computers. Such environments require a naming system that is (i) expressive, to describe and make requests based on specific properties of services, (ii) responsive, to track changes due to mobility and performance, (iii) robust, to handle failures, and (iv) easily configurable. INS uses a simple language based on attributes and values for its names. Applications use the language to describe what they are looking for (i.e., their intent), not where to find things (i.e., not hostnames). INS implements a late binding mechanism that integrates name resolution and message routing, enabling clients to continue communicating with end-nodes even if the name-to-address mappings change while a session is in progress. INS resolvers self-configure to form an application-level overlay network, which they use to discover new services, perform late binding, and maintain weak consistency of names using soft-state name exchanges and updates. We analyze the performance of the INS algorithms and protocols, present measurements of a Java-based implementation, and describe three applications we have implemented that demonstrate the feasibility and utility of INS.

High-performance execution in distributed computing environments often requires careful selection and configuration not only of computers, networks, and other resources but also of the protocols and algorithms used by applications. Selection and configuration in turn require access to accurate, up-to-date information on the structure and state of available resources. Unfortunately, no standard mechanism exists for organizing or accessing such information. Consequently, different tools and applications adopt ad hoc mechanisms, or they compromise their portability and performance by using default configurations. We propose a solution to this problem: a Metacomputing Directory Service that provides efficient and scalable access to diverse, dynamic, and distributed information about resource structure and state. We define an extensible data model to represent the information required for distributed computing, and we present a scalable, high-performance, distributed implementation. The dat...

We analyze the growth of a medium-sized research laboratory 's wide-area TCP connections over a period of more than two years. Our data consisted of six month-long traces of all TCP connections made between the site and the rest of the world. We find that smtp, ftp, and X11 traffic all exhibited exponential growth in the number of connections and bytes transferred, at rates significantly greater than that at which the site's overall computing resources grew; that individual users increasingly affected the site's traffic profile by making wide-area connections from background scripts; that the proportion of local computers participating in wide-area traffic outpaces the site's overall growth; that use of the network by individual computers appears to be constant for some protocols (telnet) and growing exponentially for others (ftp, smtp); and that wide-area traffic geography is diverse and dynamic. 1 Introduction To properly design future networks, we need a thorough understanding of...

HARNESS (Heterogeneous Adaptable Reconfigurable Networked SystemS) is an experimental metacomputing system[22] built around the services of a highly customizable and reconfigurable distributed virtual machine (DVM). The successful experience of the HARNESS design team with the Parallel Virtual Machine (PVM) project has taught us both the features which make the DVM model so valuable to parallel programmers and the limitations imposed by the PVM design. HARNESS seeks to remove some of those limitations by taking a totally different approach to creating and modifying a DVM. Keywords: metacomputing, message-passing libraries, distributed applications, distributed virtual machines, PVM 1 Introduction Virtual machine (VM) terminology, borrowed from PVM[8], refers to the fact that the computing resources on which a system runs can be viewed as a single large distributed memory computer. The virtual machine is a software abstraction of a distributed computing platform consisting of a set of ...

SNIPE is a metacomputing system that aims to provide a reliable, secure, fault-tolerant environment for long-term distributed computing applications and data stores across the global InterNet. This system combines global naming and replication of both processing and data to support large scale information processing applications leading to better availablity and reliability than currently available with typical cluster computing and/or distributed computer environments. Keywords: SNIPE, RCDS, MetaComputing, scalable, secure, reliable Acknowledgements This work was supported in part by the Office of Scientific Computing, U.S. Department of Energy, under Contract DE-AC05-96OR22464, by DARPA under Contract DAAH 04-95-1-0595, and by the National Science Foundation's Center for Research on Parallel Computation, Science and Technology Center Cooperative Agreement No. CCR-8809615. 1. Introduction The beginning of the 21st century will present new challenges for large-scale applications i...

this paper takes both a look back and a look forward to describe research directions for this important multimedia application area.

With mobile computing devices and services becoming more prevalent, the need is growing for mechanisms in a network that allow higher-level service interaction. Network-layer connectivity is not sufficient for devices to discover and understand relevant services in their vicinity; for example, simply being connected to the network does not allow a device to find the least-loaded printer or locate the nearest web camera. Several projects and systems related to the problem of service discovery are underway, including Sun's Jini, Berkeley's Service Discovery System, and the Intentional Naming System (INS) being developed at MIT LCS. In the Intentional Naming System, scalability limitations arise when trying to scale beyond the scope of a single local organization. This comes mainly from the assumption, also made by its counterparts, that a server is able to know about all services in the world, and it is furthermore necessitated by the lack of scalability-related structure in its names. To al...

With the growth of the Internet as a whole, and the popularity of the World-Wide Web information system in particular, Uniform Resource Locators (URLs) have become the de facto standard for naming on-line resources. The usage of URLs on a large scale has highlighted their deficiences, and new technologies are being developed which attempt to address these - notably the Uniform Resource Name (URN) and Uniform Resource Characteristic (URC). This report outlines some of the problems with URLs, provides an introduction to URNs and URCs, and identifies areas where the approach adopted for the Simple Discovery Protocol may be a viable solution. 1 Preamble It is assumed that the reader is already familiar with the Internet, Uniform Resource Locators[1], and the World-Wide Web. There are now a large number of elementary texts which introduce these concepts, for example[2]. This report discusses problems arising from the current usage of URLs on the Internet, and introduces some possible so...

this paper, we develop a framework for a modern hierarchical data model, substantially improved from the original version by taking advantage of the lessons learned in the relational database context. We argue that this new hierarchical data model has many benefits with respect to the ubiquitous flat relational data model.

Lists of entities must often be specified in many real-world applications such as customer lists, electronic distribution lists and access control lists. These lists are typically specified through explicit enumeration, frequently aided by recursive expansion. In this paper, we discuss the declarative specification and extraction of members of such lists as queries over a directory that maintains information both about individuals and about lists, and identify key features that the directory must support to manage lists in a flexible manner. X.500 is the industry standard for modeling information about individuals in a directory, and LDAP is the proposed standard for accessing directory information. We have designed and built a system to represent and manage lists in the X.500 information model, and developed efficiently evaluable extensions to the LDAP query language for the location and expansion of lists. We describe the system architecture and the query evaluation algorithm of this...