Showing papers on "Edge computing published in 2006"

A Scalable Cluster-based Infrastructure for Edge-computing Services

Abstract: In this paper we present a scalable and dynamic intermediary infrastructure, SEcS (acronym of "Scalable Edge computing Services"), for developing and deploying advanced Edge computing services, by using a cluster of heterogeneous machines. Our goal is to address the challenges of the next-generation Internet services: scalability, high availability, fault-tolerance and robustness, as well as programmability and quick prototyping. The system is written in Java and is based on IBM's Web Based Intermediaries (WBI) [71] developed at IBM Almaden Research Center.

Three architectures for trusted data dissemination in edge computing

Abstract: Edge computing pushes application logic and the underlying data to the edge of the network, with the aim of improving availability and scalability. As the edge servers are not necessarily secure, there must be provisions for users to validate the results--that values in the result tuples are not tampered with, that no qualifying data are left out, that no spurious tuples are introduced, and that a query result is not actually the output from a different query. This paper aims to address the challenges of ensuring data integrity in edge computing. We study three schemes that enable users to check the correctness of query results produced by the edge servers. Two of the schemes are our original contributions, while the third is an adaptation of existing work. Our study shows that each scheme offers different security features, and imposes different demands on the edge servers, user machines, and interconnecting network. In other words, all three schemes are useful for different application requirements and resource configurations.

Autonomic management of edge servers

Abstract: Delivering innovative Internet services raises numerous business and technical challenges for providers. It actually requires building and managing complex, distributed architectures in order to reach the quality of service that is needed. In this paper, we argue that using edge computing in the domain of Internet services has a number of advantages. However, this approach relies on complex and hard to administrate environments. We believe that autonomic computing techniques constitute a key element for the dynamic management of edge servers. In the paper, we present an autonomic manager that meets the market needs and that has been tested in collaboration with Bull SA.