Igor Bychkov

Multiagent approach to controlling distributed computing in a cluster Grid system

Generally, distributed computing environments have a number of properties that significantly complicate unification of processes of computing control such as their scheduling and resource allocation. Such properties include, for instance, functional-organizing heterogeneity, dynamics and non-complete description of integrated resources; diversity of the spectrum of problems solved using these resources; different classes of users pursuing their own goals and tasks while working with the computing system. Analysis of world trends in this research domain allows stating that solving these problems is directly connected with intellectualization of middleware of distributed computing environments with decentralized control. In this work, a multiagent approach to controlling distributed computing in a cluster Grid system, which is a virtual software-hardware infrastructure with its nodes being computer clusters, is presented. Characteristics of such system are considered. Architecture and principles of operation of the multiagent system are given. A number of important technological features of the proposed approach is singled out. The multiagent system is developed using JADE tools (Java Agent DEvelopment framework). The results of imitation simulation of the processes of operation of the system of agents being developed are given.

Agent-Based Approach to Monitoring and Control of Distributed Computing Environment

This paper discusses a problem of monitoring heterogeneous distributed computing environments which consist of loosely coupled multiplatform computing resources. We propose an approach to the organization of the meta-monitoring system which collects data from existing local monitoring systems and own software sensors, unifies and analyzes data, generates necessary control actions. The approach is based on web-technologies, multi-agent technologies, expert systems, methods of decentralized processing and distributed storage of data.

Service-oriented multiagent control of distributed computations

Consideration was given to the multiagent methods and toolkits for efficient control of the job flows generated by the service-oriented applications. These designs were integrated within the framework of a unique technology supporting automation of solution of large scientific problems in the up-to-date cluster Grid whose computing nodes (clusters) can be of an involved hybrid structure. The novelty and practical significance of the methods and tools described in the paper lie in essential extension of the functionality of the computation control system of the cluster Grid, as compared with the existing ones, distribution and sharing of the Grid resources at various levels of job execution, and possibility of integrating intelligent computation control tools in the problem-oriented applications.

Multiagent control of computational systems on the basis of meta-monitoring and imitational simulation

This paper describes the control of computations in a distributed computing environment (DCE) on the basis of its meta-monitoring and simulation modeling. Computations are controlled by a multiagent system with a given organizational structure. Resource allocation is carried out by agents with the use of economic mechanisms for controlling their supply and demand. Controlling actions for agents are formed on the basis of the simulation modeling of functional processes of the DCE. Data about the DCE resources and processes are collected and emergency situations in the DCE nodes are detected and prevented by the meta-monitoring system of this environment. The research results are the techniques for selecting control actions and the methods for intellectual processing and effective storage of data.

A Volunteer-Computing-Based Grid Architecture Incorporating Idle Resources of Computational Clusters.

In this paper, we suggest a new architecture of a computational grid that involves resources of BOINC-based volunteer computing projects and idle resources of computational clusters. We constructed a computational grid of the proposed kind, based on several computational clusters and the volunteer computing project SAT@home. This project, launched and maintained by us, is aimed at solving hard computational problems, which can be effectively reduced to Boolean satisfiability problem. In the constructed grid several new combinatorial designs based on diagonal Latin squares of order 10 were found, and also several weakened cryptanalysis problems for the Bivium cipher were solved.

Hybrid Evolutionary Approach to Multi-objective Mission Planning for Group of Underwater Robots

We propose a hybrid approach, based on the combined use of genetic algorithms, methods and heuristics of local search, and ant colony optimization to solve the dynamic routing problem for a group of underwater robots. Group’s objective involves visiting a certain set of control points (for the purpose of sampling, taking measurements, photos and videos) according to their priority and under given restrictions. The dynamic routing problem here is to find (planning) and adjust (replanning) feasible group routes for robots, ensuring as far as possible the maximum efficiency of the group work.

Concept of a multitask grid system with a flexible allocation of idle computational resources of supercomputers

A new concept of a multitask distributed heterogeneous computing system is proposed. The basic principles of such system are that it uses only idle supercomputer resources and does it as a common user; thus, it does not conflict with the administration policy in any way. The efficiency of the proposed concept is demonstrated by the example of the real grid system that currently uses supercomputer resources to boost the performance of the SAT@home and OPTIMA@home volunteer distributed computing projects.

An intelligent management system for the development of a regional transport logistics infrastructure

This paper presents an intelligent management system realized on the basis of ontologies and formalized expert knowledge, mathematical models and numerical methods. We introduce a complex approach to the analysis of transport logistics systems using an original concept of multilevel modeling. The constructed mathematical models and software modules integrated in the system are described in brief. Modern intellectualization tools are applied for providing interaction among different modules and processing of incomplete statistical data. And finally, we consider some test and applied problems and their solution by the system.

On the problem of discrete-event systems properties preservation

The paper presents a novel approach to solving a problem generally arising in studying dynamical systems, namely the problem of a system’s properties preservation under some transformation. Combining algebra, logic and dynamics, the method of logical-algebraic equations (LAE-method) is developed, serving to synthesize criteria for preservation properties of systems connected by special type of morphisms. The LAE-method is applicable to various systems, but we focus on the case of discrete-event systems (DES), which are the systems that evolve in time due to the occurrence of some event sequences. We consider the issues of the LAE-method application to the reduction of supervisor for DES, the problems of DES basic properties, such as observability and controllability, preservation when sensor readings provide information about system’s state and it is available to a supervisor. Decentralized supervisory control is also addressed, in particular, the question whether local supervisors properties are inherite...

A hybrid approach to solve the dynamic patrol routing problem for group of underwater robots

We suggest an approach to solve the dynamic patrolling problem as the variation of periodic vehicle routing problem with time windows. In order to do this, firstly a set of waypoints (targets) should be located in space in such way that their spherical neighborhoods, which corresponds to the size of the robots vision/detections fields, would form a complete coverage of the protected perimeter. Regular inspections of each waypoint by robots of the group not less than once during specified period would guarantee the effective perimeter patrolling. Thus, the routing problem is to a feasible group route ensuring well-timed inspections for all waypoints of the mission. To solve the described problem we propose a hybrid approach, based on the combined use of genetic algorithms and methods and heuristics of local search. The proposed approach allows robots to effectively generate required solutions under given constraints and quickly readjust it in unforeseen situations due to the dynamic nature of the problem. A software modeling system implementing all the necessary computational procedures has been developed; the results of computations are given.