We investigate the problem of spreading information contents in a wireless ad hoc network with mechanisms embracing the peer-to-peer paradigm. In our vision, information dissemination should satisfy the following requirements: (i) it conforms to a predefined distribution and (ii) it is evenly and fairly carried by all nodes in their turn. In this paper, we observe the dissemination effects when the information moves across nodes according to two well-known mobility models, namely random walk and random direction. Our approach is fully distributed and comes at a very low cost in terms of protocol overhead; in addition, simulation results show that the proposed solution can achieve the aforementioned goals under different network scenarios, provided that a sufficient number of information replicas are injected into the network. This observation calls for a further step: in the realistic case where the user content demand varies over time, we need a content replication/drop strategy to adapt the number of information replicas to the changes in the information query rate. We therefore devise a distributed, lightweight scheme that performs efficiently in a variety of scenarios.
Deep Dive into A Holistic Approach to Information Distribution in Ad Hoc Networks.
We investigate the problem of spreading information contents in a wireless ad hoc network with mechanisms embracing the peer-to-peer paradigm. In our vision, information dissemination should satisfy the following requirements: (i) it conforms to a predefined distribution and (ii) it is evenly and fairly carried by all nodes in their turn. In this paper, we observe the dissemination effects when the information moves across nodes according to two well-known mobility models, namely random walk and random direction. Our approach is fully distributed and comes at a very low cost in terms of protocol overhead; in addition, simulation results show that the proposed solution can achieve the aforementioned goals under different network scenarios, provided that a sufficient number of information replicas are injected into the network. This observation calls for a further step: in the realistic case where the user content demand varies over time, we need a content replication/drop strategy to ad
It is commonly acknowledged that user devices are rapidly becoming tantamount to a communication hub, sporting arrays of GPS navigators, web browsers, videogame consoles and screens flashing the latest news or local sightseeing information. In this context, most pieces of information are likely to be of general use, and therefore a sensible dissemination and caching policy would be desirable.
In this work, we focus on such an environment: few and far between access points, or gateway nodes, in a highlypopulated network area where user devices are equipped with a data cache and communicate through the ad hoc networking paradigm. Users create a cooperative environment where information is exchanged among nodes in a peer-to-peer fashion. In particular, they form a pure peer-to-peer system, whose nodes may simultaneously act as both “clients” and “servers” to the other nodes in the network. Also, we envision a system that achieves a desired distribution over the network of the information that users may be interested in. By information distribution, we mean the distribution according to which the information copies should be cached in the network. The nodes storing an information copy will act as providers for this content.
Traditional approaches to information caching in communication networks [1]- [4] are based on the solution of linear programming problems, which often require global knowledge of the network condition, or lead to quite complex solutions that involve significant communication overhead. Unlike previous approaches, our solution is fully distributed and it comes at a very low cost in terms of communication overhead. Our goal is to achieve the desired content distribution by properly letting the information move across the network.
More specifically, while developing our solution, we identify a number of issues that need to be addressed.
• Achieving the desired distribution of the information: regardless of how the information is distributed at the outset, the system should be able to identify where the information should be stored in the network.
• Fair distribution of information burden: as mentioned above, a node storing the information acts as provider for that information; of course, this role may exact a high toll from nodal resources in terms of bandwidth or power consumption; it is therefore advisable that the role of content provider be handed over to neighboring nodes quite frequently, without altering the information distribution.
• Dynamic adaptation to time-varying content demands:
given an initial number of information replicas, which implies a fair load distribution among the provider nodes, the system should be able to adapt to the varying information query rate, by increasing or decreasing the number of information copies in the network as needed. We deal with the above issues and develop a solution that features the following advantages:
• it is fully distributed;
• it is content-transparent, i.e., it does not require knowledge of the contents stored by the neighboring users;
• it works with minimum overhead.
In particular, motivated by the need of a balanced load distribution among the provider nodes and of an equal quality of service provisioning to the users, we target a uniform distribution of contents, either over the network spatial area or over the network nodes. With this aim in mind, in Section III we investigate the applicability of two well-known mobility models, namely the random walk and the random direction model, to disseminate the information across the network. Both strategies, using the experimental setup in Section IV, are proven to yield a distribution of the information copies that is close to the target distribution, regardless of the considered network scenario (Section V). Also, the obtained results show that the level of fairness in distributing the burden among provider nodes depends on the number of information copies stored in the network. Thus, when the user query rate for a content varies over time, we address the problem of how to let the number of content replicas adapt to the changing content demand. To address this issue, in Section VI we devise a content replication/drop algorithm that controls the number of information copies in the network, taking into account the time-varying behavior of the contents popularity level. The performance of our scheme is presented in Section VI-A.
Our study is related to the problem of optimal cache placement in wireless networks. Several works have addressed this issue by exploiting its similarity to the facility location and the k-median problems. Both these problems are NP-hard and a number of constant-factor approximation algorithms have been proposed for each of them [1], [3], [5]; these algorithms however are not amenable to an efficient distributed implementation.
Distributed algorithms for allocation of information replicas are proposed, among others, in [4], [6]- [8]. These solutions typica
…(Full text truncated)…
This content is AI-processed based on ArXiv data.
