ADAPTIVE SECURITY POLICY FOR  DYNAMIC SOFTWARE COMPONENT ADAPTATION IN  DISTRIBUTED SYSTEM ARCHITECTURE

MINASIE, LEMMA

dc.contributor.author	MINASIE, LEMMA
dc.date.accessioned	2022-11-17T11:55:56Z
dc.date.available	2022-11-17T11:55:56Z
dc.date.issued	2022-03
dc.identifier.uri	http://ir.bdu.edu.et/handle/123456789/14447
dc.description.abstract	A distributed system is a group of independent computers that, from the perspective of a user, operate as a single system. Individual computers may have software components that migrate from one to another, add or remove themselves dynamically, or be connected or disconnected with them. Security becomes a serious concern when a distrusted system component has the ability to freely configure itself. Security concerns can be classified into two categories: reallocation of software components to a malicious host and relocation of malicious software components to the host or execution of a software components program on hosts. The aims of this study is to develop and evaluate adaptive security policy for dynamic software component adaptation approach in distributed system architecture. In this study, the simulation research method was used. It comprises choosing software and host security level us parameters, establishing the experiment environment, and evaluating software component adaptation based on the parameters we selected. This enabled us to identify security levels impacted by dynamic adaptation in terms of percentage of security violations in each adaption, which ranged from 43 percent to 75 percent of adaptation of software components to an unprivileged host and unprivileged software components to the host in 10 different simulations. This study mitigated adaptation of software components and hosts to unprivileged hosts and software components, as indicated in simulation and evaluation findings, by using dynamically generated lattices of security levels. To generate lattices dynamically, we developed adaptive security policies by adapting extended Dijkstra's guarded command language with mathematics concepts such as partially ordered set, lattice, and incidence matrix. Based on simulation and evaluation results, we conclude that, if the security domains of the software component and the host can be partially ordered, we can protect the software component and the host from adapting in malicious host and software component by generating a dynamic security lattice that meets proposed adaptive security policy. Key words: Dynamic Adaptation, Adaptive Security Policy, Software Component, Distributed System, Security Lattice, Partially Ordered Set, Incidence Matrix	en_US
dc.language.iso	en_US	en_US
dc.subject	FACULTY OF COMPUTING	en_US
dc.title	ADAPTIVE SECURITY POLICY FOR DYNAMIC SOFTWARE COMPONENT ADAPTATION IN DISTRIBUTED SYSTEM ARCHITECTURE	en_US
dc.type	Thesis	en_US