Today's universal communications increasingly involve mobile and battery-powered devices (e.g. hand-held, laptop) over wired and wireless networks. Energy efficiency as well as throughput, are becoming service characteristics of dominant importance in communication protocols. Although standard TCP versions lack the functionality to efficiently adjust their error-control strategies to distinct characteristics of network environments and to specific constraints of communicating devices, the wide range of TCP-based applications have rendered TCP the de facto standard for reliable end-to-end communications. In this work we propose "grafting" two components of strategic significance onto standard TCP: a probing mechanism and an immediate recovery strategy. Our results show that these enhancements yield higher throughput while maintaining lower levels of energy expenditure, and thus have the potential of promoting TCP's congestion control to a universal error control schema for heterogeneous wired/wireless channels. Furthermore, the enhancements do not damage the end-to-end characteristics of the TCP, nor do they require changes to its semantics: the mechanism are implemented as option extensions to the TCP header. We compare "TCP-probing" with Tahoe, Reno, and New Reno, and show that is can be a protocol of choice for heterogeneous wired/wireless communications with respect to energy and throughput performance.

Today's universal communications increasing involve mobile and battery-powered devices (e.g. hand-held, laptop, IP-phone) connected over wired and wireless networks. Energy efficiency, as well as throughput are becoming service characteristics of dominant importance in communication protocols. The wide applicability of IP-networks/devices, and the wide range of TCP-based applications, have rendered TCP the de facto reliable transport protocol standard not only for wired, but also for wireless and mixed (wired/wireless) communications. TCP's congestion control algorithms have been refined to achieve higher throughput. Even with these modifications, however, TCP versions do not incorporate a flexible error recovery strategy that is responsive to distinct environmental characteristics and device constraints. We have compared the energy- and throughput-efficiency of TCP error control strategies based on results gathered from our implementation of TCP Tahoe, Reno, and New Reno. We show that, depending on the frequency and duration of the error, each demonstrates appropriate behavior under specific circumstances, with Tahoe more-or-less the most energy conserving of the three. None of them, however, possesses a clear-cut overall advantage that would render it the version of choice for wired/wireless heterogeneous networks.

This work involves the development of an experimental transport-level protocol running on top of IP for small, mobile, wireless stations. The protocol cannot and does not aim to handle real-time traffic. Its central concern is to conserve battery-powered energy used for transmission, even at the expense of slower data throughput. It attempts to conserve energy expenditure by adjusting the amount of data transmitted, trying to keep it below perceived network congestion. The higher the detected congestion risk-level in the network the less it attempts to transmit thereby minimizing the need for duplicate data retransmission due to congested routers losing packets and so on. We outline the protocol specification, mechanisms and implementation, as well as preliminary evaluation results that clearly demonstrate the energy-saving capabilities of the protocol.

With the advent of distributed computing, application management is driven into a new era. In this world, application quality of service, resource availability at different domains, and user requirements, are combined in a dynamic, synergistic, and geographically broad cooperation. We present an architecture and initial implementation of a middleware that satisfies user requests according to expected requirements, distributes networked resource load beyond the local environment, and respects the logical constraints of different domains. Static characteristics of resources are advertised as service offers using the CORBA trading service, while dynamic changes in their state are reported using the CORBA event service. We present a model where, unlike traditional high-level management systems, resources compete to satisfy application requirements.

A probabilistic routing policy, the Z/sup 2/ (zigzag) routing policy, is presented within the class of nonadaptive, shortest-path routing policies for regular mesh-connected topologies such as n-dimensional toroids and hypercubes. The focus of the research is routing in networks of computers in a distributed computing environment, where constituent subcomputers are organized in a mesh-connected topology and communication among individual computers takes places by some form of message exchange. The authors prove the optimality of this policy with respect to two criteria: (1) maximizing the probability of reaching the destination from a given source without delays at intermediate nodes; and (2) minimizing the expected lifetime of a message.>

A network computer is a collection of computers designed to function as one machine. On a network computer, as opposed to a multiprocessor, constituent subcomputers are memory-disjoint and communicate only by some form of message exchange. Ensemble architectures like multiprocessors and network computers are of growing interest because of their capacity to support parallel programs, where a parallel program is one that is made up of many simultaneously-active, communicating processes. Parallel programs should, on an appropriate architecture, run faster than sequential programs, and, indeed, good speed-ups have been reported in parallel programming experiments in several domains, amongst which are AI, numerical problems, and system simulation. Our interest lies in network computers, particularly ones that range in size from several hundred nodes to several thousand.

Staged circuit switching (SCS) is a message-switching technique that combines a new protocol with new communication hardware. Protocol and hardware are designed specifically for networks that are intended to function as integrated general-purpose MIMD machines, i.e., for "network computers."

Staged circuit switching (SCS) is a message-switching technique that combines a new protocol with new communication hardware. Protocol and hardware are designed specifically for networks that are intended to function as integrated, general-purpose MIMD machines, i.e. for "network computers".