BEHAVIOUR-BASED TRUST MODEL IN A GRID COMPUTING SYSTEM BASED ON FUZZY LOGIC

Abstract

This paper is dedicated to exploring the application fuzzy logic for the development and improvement of a trust model for computational elements in grid computing systems. Such systems have gained significant popularity due to their flexibility, scalability and ability to efficiency incorporate new nodes, thereby increasing overall computational power available for solving complex problems. The effectiveness of grid systems has been successfully demonstrated through numerous scientific, academic and domain-specific projects that involve a large number of users voluntarily contributing their own resources to perform computationally intensive tasks. However, despite these clear advantages, the challenge of ensuring a high level of security, reliability and accuracy of computational results remains both relevant and complex. The main difficulty lies in the lack of direct control over the software and hardware components of the computing elements that join the system. As a result, there is a need for repeated verification of obtained results, which certainly increases system load and reduces overall performance. Without such verification, there is a risk of result alteration due to the specific nature of task execution in this environment. Since computation are divided into smaller parts and distributed among a large number of nodes, with partial results later aggregated into single output, even one error or malicious action can critically affect the accuracy of the final result. Given this, the goal of the article is to present a modified trust model based on fuzzy logic, which enables the appropriate distribution of roles among elements in the system. This modification is expected to reduce the number of required result verifications and improve the overall efficiency of grid system functioning.