Computational grids have not scaled effectively due to administrative hurdles to resource and user participation. Most production grids are essentially multi-site supercomputer centers, rather than truly open and heterogeneous sets of resources that can join and leave dynamically, and that can provide support for an equally dynamic set of users. Large-scale grids containing individual resources with more autonomy about when and how they join and leave will require self-organizing grid middleware services that do not require centralized administrative control. This paper considers one such service, namely the dynamic discovery of high-performance variable-size clusters of grid nodes. A brute force approach to the problem of identifying these "ad-hoc clusters" would require excessive overhead in terms of both message exchange and computation. Therefore, we propose a scalable solution that uses a delay-based overlay structure to organize nodes based on their proximity to one another, using a small number of delay experiments. This overlay can then be used to provide a variable-size set of promising candidate nodes than can then be used as a cluster, or tested further to improve the selection. Simulation results show that this approach results in effective clustering with acceptable overhead.

/pdf/automatic-clustering-for-self-organizing-grids-1sf0g2xhfx.pdf

Automatic Clustering for Self-Organizing Grids

The claimed subject matter provides a system and/or a method that facilitates integration of a distributed backup environment and a online backup environment. A super peer device can be designated from a set of peer devices. The super peer can distribute backup data amongst the set of peer devices based upon availability and storage capacity of the peer devices. In addition, the super peer can transfer portions of backup data from the set of peers to an online backup service.

Hybrid distributed and cloud backup architecture

Optimizing cell sites or sectors in a wireless network includes calculating modifications to a plurality of network parameters for optimizing wireless network performance; evaluating the modification groups to determine conflicts between modifications for a same network parameter: and eliminating the conflicts between modifications for the same network parameter within the modification groups The modification groups are used to alter at least one network parameter of the critical ceil sites or sectors, or of a best neighbor ceils sites or sectors for achieving a desired improvement in communications within the wireless network Optimization is further enhanced by determining the best previous modifications to the wireless network when performance continues to be degraded Altering wireless network parameters of the critical cell sites or sectors, or the best neighbor cell sites or sectors is performed continuously using the stored modification groups until die desired improvement in communications in the wireless network is achieved

Self-optimizing wireless network

The embodiment of the invention provides a cloud processing system, a cloud processing method, and a cloud computing agent device, as well as a cloud storing and processing device and a client, wherein the cloud processing system comprises a client, a cloud server and a cloud client device; the cloud client device is connected with the client and the cloud server; and the cloud client device is used for receiving the cloud processing request messages sent out by the client, so as to provide service for a client instead of the cloud server. The embodiment of the invention further provides the corresponding cloud processing method, the cloud computing client device and the cloud storing device. By adopting the method, the device and the system provided by the embodiment of the invention, the information safety during the cloud processing can be improved.

Cloud processing system, cloud processing method and cloud computing agent device

Disclosed are methods and apparatus that allow users and organizations to access on-demand and in a personalized way to the performance and optionally flow activity measures of an end to end network of virtual resources. The system is configured to generate a UI that provides a mapping and characterization of the network of virtual resources. In one embodiment, a network topology map can be generated in the UI. In another embodiment, a flow map including nodes and flows can be generated in the UI. The UI is configured to provide users with a number of actions that can be implemented to affect the virtual resources. The actions can be presented in the user via menu options or actionable objects. The user can implement high-level actions provided by the UI. Upon receipt of a high-level action, the system is configured to translate the action into a sequence of low-level device commands needed to implement the action. The system can communicate with the low-level devices in the cloud to implement the commands.

User interface for visualizing resource performance and managing resources in cloud or distributed systems

A distributed mediation network and method of employing such is provided, having a plurality of different types of network module. Each module has a non-reciprocal path therethrough for network traffic and the distribution of network traffic across the network is managed by an autonomic control plane.

Alex Heneveld

Papers

Self-managed distributed mediation networks