Demand-driven publish/subscribe in mobile environments

Abstract

We propose a novel routing protocol, SelfBalancing Supply/Demand (SBSD), for Publish/Subscribe in mobile ad hoc environments. SBSD is a controlled flooding that reduces network congestion by constraining how far subscriptions replicate and how many times nodes broadcast them. SBSD ranks subscriptions by a utility function. This function matches the supply of publications with the recent demand for them; more popular subscriptions are replicated farther and their replicas are retained longer. SBSD is therefore demand-driven, as more popular subscriptions are more likely to receive their matching publications and receive them sooner. We show that SBSD is scalable; routing distance is independent of network size. SBSD's performance is examined under random mobility, under challenging conditions including high node mobility and broadcast failure rates.

Probabilistic source-based query subsumption in vehicular ad hoc networks

“…Such mixedmobility networks are denser and less vulnerable to partitioning than separate networks for vehicles and pedestrians. Messages are routed using Self-Balancing Supply/Demand (SBSD) controlled flooding protocol, which is effective in dynamic topologies [4]. SBSD uses a congestion-based ranking metric called utility; packets closer to their spatio-temporal point of origin have higher utility.…”

“…Inflicted congestion is estimated from the time and distance a message has flooded from its source, increasing with both. The above utility function causes allowed congestion to scale linearly with frequency [4]. More popular queries flood farther and receive more broadcasts at each node.…”

Dynamic Subscription Permission: Extending the Depth of Demand-Controlled Flooding

“…We present dynamic subscription permission (DSP), a method for improving throughput in self-balancing supply/demand (SBSD) routing protocols [14,15], which combine the publish/subscribe paradigm [6,9] and demand-controlled flooding. DSP makes SBSD use bandwidth more efficiently in dense environments by reducing the number of nodes broadcasting the same subscriptions without materially reducing coverage.…”

“…These areas also tend to be saturated; that is, each node tends to hold a similar set of subscriptions as its neighbors. While this provides reliable publication delivery even in large, high-mobility networks [15], the propagation areas may be small when a large variety of subscriptions is active. This makes it unlikely for publications to reach distant subscribers.…”

“…Utility (u) measures how long a subscription has existed in the network, how far it has propagated, and how many nodes have made the same request. Queuing by (1) lets the total bandwidth allocated to each subscription over its lifetime scale linearly with the number of subscribers [15]. Each node only queues the highest-utility subscriptions it has received and no more than one of each type; duplicate subscriptions are handled by updating the demand field.…”