Abstrait

Decentralized Data Allocation with Load Balancing Mechanism

A. Rajalakshmi, Dr. A. M. J. Md Zubair Rahman, ME, MS, Ph.D.,

Mutual data sharing is provided in distributed file sharing environment. Information brokering systems (IBSs) are used to connect large-scale loosely federated data sources via a brokering overlay. Information brokers redirect the client queries to the requested data servers. Privacy preserving methods are used to protect the data location and data consumer. Brokers are trusted to adopt server-side access control for data confidentiality. Query and access control rules are maintained with shared data details under metadata. A Semantic-aware index mechanism is applied to route the queries based on their content and allow users to submit queries without data or server information. Privacy Preserved Information Brokering (PPIB) scheme is used to preserve privacy for distributed data sharing process. Attribute-correlation attack and inference attacks are handled by the PPIB. PPIB overlay infrastructure consisting of two types of brokering components, brokers and coordinators. The broker’s acts as mix anonymizer are responsible for user authentication and query forwarding. The coordinators concatenated in a tree structure, enforce access control and query routing based on the automata. Automata segmentation and query segment encryption schemes are used in the Privacy-preserving Query Brokering (QBroker). Automaton segmentation scheme is used to logically divide the global automaton into multiple independent segments. The query segment encryption scheme consists of the preencryption and post encryption modules. The PPIB scheme is improved to support dynamic site distribution and load balancing mechanism. Peer workloads and trust level of each peer are integrated with the site distribution process. The PPIB is improved to adopt self reconfigurable mechanism. Automated decision support system for administrators is included in the PPIB. The system reduces the data distribution overhead and response time for the clients.