Dynamic Differential Data Protection for High-Performance and Pervasive Applications

Distributed systems;Security;Publish-subscribe systems;Middleware;Data protection

Widener, Patrick M. (Patrick McCall) ; Computing

University:Georgia Institute of Technology

Department:Computing

关键词: Distributed systems;    Security;    Publish-subscribe systems;    Middleware;    Data protection;   

Others  :  https://smartech.gatech.edu/bitstream/1853/7239/1/widener_patrick_m_200508_phd.pdf

美国|英语

来源: SMARTech Repository

【 摘 要 】

Modern distributed applications are long-lived, are expected toprovide flexible and adaptive data services, and must meet thefunctionality and scalability challenges posed by dynamically changinguser communities in heterogeneous execution environments.Thepractical implications of these requirements are that reconfigurationand upgrades are increasingly necessary, but opportunities to performsuch tasks offline are greatly reduced.Developers are responding tothis situation by dynamically extending or adjusting applicationfunctionality and by tuning application performance, a typical methodbeing the incorporation of client- or context-specific code intoapplications' execution loops.Our work addresses a basic roadblock in deploying such solutions: the protection of keyapplication components and sensitive data in distributed applications.Our approach, termed Dynamic Differential Data Protection (D3P),provides fine-grain methods for providing component-based protectionin distributed applications. Context-sensitive, application-specificsecurity methods are deployed at runtime to enforce restrictions indata access and manipulation. D3P is suitable for low- orzero-downtime environments, since deployments are performed whileapplications run.D3P is appropriate for high performance environmentsand for highly scalable applications like publish/subscribe, becauseit creates native codes via dynamic binary code generation. Finally,due to its integration into middleware, D3P can run across a widevariety of operating system and machine platforms.This dissertation introduces D3P, using sampleapplications from the high performance and pervasive computing domainsto illustrate the problems addressed by our D3P solution. It alsodescribes how D3P can be integrated into modern middleware.Wepresent experimental evaluations which demonstrate the fine-grainnature of D3P, that is, its ability to capture individual end users'or components' needs for data protection, and also describe theperformance implications of using D3P in data-intensive applications.