Data Partitioning and Task Management in the Clustered Server Layer of the Volunteer-based Computation System

While the typical volunteer-based distributed computing system [1] focus on the computing performance, the Comcute system [3] was designed especially to keep alive in the emergency situations. This means that designers had to take into account not only performance, but the safety of calculations as well. Quadruple-layered architecture was proposed to separate the untrusted components from the core of the system. The main layer (W) consists of independent server nodes, which are coupled into a cluster. The W-servers provide task promotion among the nodes, data partitioning, results gathering, comparing and merging. The cluster remains operational as long as one of the nodes is able to operate. This paper describes the functionality of the Comcute system from the W-node perspective considering two task parameters: required performance level and required level of computational reliability.

On Configurability of Distributed Volunteer-Based Computing in the Comcute System

The chapter proposes additional solutions that can be implemented within the Comcute system to increase its configurability. This refers to configuration of the reliability level in the W and S server layers, static or on-the-fly data partitioning and integration, configuration of the system for processing in the data streaming fashion, extending the system for selection of a project that the client wants to contribute to, ease of migration of legacy codes to the system. Finally, an example of a legacy distributed application for monitoring client locations and resource usage is presented with suggestions on its migration to the Comcute system environment.

Foundations of Grid Processing Architecture for the Comcute System

In the chapter, fundamental system algorithms and structures implemented in the Comcute system are described and analysed in detail. Layered architecture of the system model is highlighted. System tasks of the layers are elaborated, presented and described. Operational details of communication interfaces among layers are worked out and examined. The focus is put onto implemented system components with regard to their operability and efficiency. Scalability and system openness, as the key design factors of the implementation, are deliberatively taken into account. Important aspects of operability are addressed and the issues of validation, verification and deployment of the adopted system solutions are discussed. Practical application aspects of the Comcute system are described with respect to its final implementation and target installation in the form of grid processing in the open worldwide Internet.

Multi Agent Grid Systems

This chapter presents an idea of merging grid and volunteer systems with multi agent systems. It gives some basics concerning multi agent system and the most followed standard. Some deliberations concerning such an existing systems were made in order to finally present possibilities of introducing agents into the Comcute system.