Ivan Lirkov

Learn More
Recently we have proposed an approach to utilizing agent teams as resource brokers and managers in the Grid. Thus far we have discussed the general overview of the proposed system, how to efficiently implement matchmaking services, as well as proposed a way by which agents select a team that will execute their job. In this paper we focus our attention on(More)
The recently introduced circulant block{factorization preconditioners are studied. The general approach is rst formulated for the case of block tridi-agonal sparse matrices. Then an estimate of the relative condition number for a model anisotropic Dirichlet boundary value problem is derived in the form < p 2"(n + 1) + 2, where N = n 2 is the size of the(More)
The Agents in Grid project is devoted to the development of an agent-based intelligent high-level Grid middleware. In the proposed system, all data processing is ontology-driven, and initially was based on an in-house developed mini-ontology of the Grid. Our recent analysis has indicated that we should adapt and utilize the Grid ontology developed within(More)
Numerical homogenization is used for up-scaling of a linear elasticity tensor of strongly heterogeneous micro-structures. Utilized approach assumes presence of a periodic micro-structure and thus periodic boundary conditions. Rotated trilinear Rannacher-Turek finite elements are used for the discretization, while a parallel PCG method is used to solve(More)
We are developing an agent-based intelligent middleware for the Grid. It is based on agent teams as resource brokers and managers. Our earlier work resulted in a prototype implementation. However, our recent research led to a complete redesign of the system. Here, we discuss the new and main technical issues found during its implementation.