Atsuji Sekiguchi

An Efficient Failure Recovery Scheme for Next Generation Network Services Based on Distributed Components

With advances of information and communication technologies, many network services can be provided by assortment of desired components running on some nodes in the network, like a distributed component based service. It is effective to reduce the cost of service development. In addition, when a failure occurs, a distributed component based service has multiple components of each function, which can be used to replace the failed component. There is another problem, however a component failure affects the application servers using it. Therefore, in this paper, we propose a novel failure recovery scheme that constructs overlay networks between components having the same function and that provides an efficient search method for the substitute component using these multiple overlay networks. Finally, simulation experiments demonstrate the excellent performance of the proposed scheme.

Multilayer failure detection method for network services based on distributed components

Network services based on distributed components, which provide advanced network services with lower cost by assigning and reusing useful components running on remote nodes, are attracting increasing attention. In this service, when a component fails, a component that has the same function running on another node can be substituted for it. On the other hand, there are three kinds of failures in network services; a software failure at a component, a hardware failure at a node, and a failure of the physical network. We propose a new method to identify a failure in network services, which determines not only a failure at the application level but also in the network layer by collecting a small number of messages via multiple overlay networks.

Dynamic reconstruction of multiple overlay network for next generation network services with distributed components

Because of rapid improvement in information and network technologies, many kinds of network services are created. In this paper, the concept of a new service platform is presented. Services are organized by components, and replications of components are distributed to many computers. Thanks to these replications, users can continue services by using another component as a substitute in case of failure. It is necessary to ask another component to fnd a substitute when a component fails. However, depending on which component to ask, the original version of the proposed method has a possibility to find only low-performance components. It is caused by constructing overlay network only once and keeping it as it is. We propose a new dynamic reconstruction method for overlay networks. Simulation experiments show how proposed method improves performance of our framework.