1,721,276 research outputs found
Publish/subscribe on the top of overlay networks: When subscription dynamics meet churn
No full textDistributed event routing on the top of overlay networks has emerged as a key technology for achieving scalable information dissemination. In particular it has been used as preferential communication backbone within publish/subscribe communication system. Its aim is to reduce the network and computational overhead per event diffusion to a set (possibly large) of interested recipients despite churn and subscription changes. This talk points out the functional decomposition between event-based routing layer, the overlay infrastructure layer (structured and unstructured) and the network protocols layer. Hence the talk surveys solutions for event based routing on overlay infrastructures discussing their inherent limitations and constraints related to both the process of arrival and departure of nodes from the overlay network and the process of subscription changes at the application level
On Event Routing in Publish/Subscribe through Dynamic Networks
No full textContent-based publish/subscribe communication systems are a popular technology for many-to-many information diffusion over large scale networks. Scalable solutions are obtained considering a network of distributed event brokers, dispatching information (events) from producers (publishers) to consumers (subscribers). Many scalable and efficient solution for routing events in content-based systems exist, that selectively send events only toward the interested subscribers. However, if subscribers having similar interest are present in different parts of the network, the benefits of such routing strategies significantly decrease. In this paper we propose a novel approach for enhancing content-based routing, based on the dynamic reconfiguration of the broker network. The reconfiguration aims at aiding the routing process by placing close to each others brokers that manage subscribers with similar interests. Metrics for measuring similarity of interests are discussed and a reconfiguration algorithm is presented
Exploiting intra-object dependencies in parallel simulation
No full textThis paper introduces the notion of weak causality that models the intra-object parallelism in parallel discrete event simulation. In this setting, a run where events are executed at each object according to their timestamp is a correct run. The weak causality relation allows to define the largest subset of all runs of a simulation that are equivalent to the timestamp-based run. Finally, we describe an application of weak causality to optimistic synchronization (Time Warp) by introducing a synchronization protocol that reduces the number of rollbacks and their extent. (C) 1999 Elsevier Science B.V. All rights reserved
A caching scheme for routing in mobile ad hoc networks and its application to ZRP
No full textA large class of routing protocols for MANETs, namely, reactive protocols, employ some form of caching to reduce the number of route discoveries. The simplest form of caching is based on associating a timeout with each cache entry. Such timer-based cache schemes can increase the protocol efficiency. However, if the timeout is not well-tuned, a severe performance degradation arises as entries are removed either too early or too late from the cache. In this paper, we address the problem of designing a proactive cache scheme that does not rely on any timer-based mechanism. This scheme guarantees that valid cached routes are never removed while stale routes are removed aggressively. This proactive cache scheme has been embedded in the Zone Routing Protocol (ZRP) framework and evaluated by an extensive simulation study
A Class of High-Performance Maekawa-type Algorithms for Distributed Systems under Heavy Demand
No full textFundamentals of Distributed Computing: A Practical Tour of Vector Clock Systems
No full textThe fundamentals of distributive computing which consists of a set of processes that cooperate to achieve a common goal are discussed. The processes do not share a common global memory and communicate only by exchanging messages over a network. It is found that a key concept of asynchronous distributed system is causality. The basic vector clock properties, mechanisms, and application examples to solve the casuality problems are also presented
- …
