Middleware-based fault recovery technique for replicated DRTDBS. (24th November 2020)

• Record Type: Journal Article
• Authors: Shrivastava, Pratik; Shanker, Udai
• Is Part Of: International journal of intelligent information and database systems. Volume 14:Number 1(2021)
• Page Start: 67
• Page End: 95
• Publication Date: 2020-11-24
• Subjects: middleware -- recovery protocol -- replication protocol -- RDRTDBS; Database management -- Computer programs -- Periodicals; Information retrieval -- Computer programs -- Periodicals; Information storage and retrieval systems -- Computer programs -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Intelligent agents (Computer software) -- Periodicals; 006.33
• Languages: English
• ISSNs: 1751-5858

Abstract

Replicated distributed real time database system (RDRTDBS) is widely used in various real time applications such as financial services, travel/aviation systems, government services, autopilot systems, etc. These real time applications require ceaseless assistance from the back-end database 24 × 7. However, it is not practical to avoid all kinds of failure in the RDRTDBS. The conventional recovery algorithms prevent the processing of admitted real time transactions (RTTs) during recovery of the recovering failed site. Our objective is to propose the recovery algorithm that simultaneously allows the processing of admitted RTTs and recovery of recovering failed replica site. In the current paper, middleware has been proposed for recovering the abandoned replica site and synchronising the consistency among replica sites. The recovery is conducted explicitly by the feedback sublayer, which consists of four algorithms: status collector, suspect confirmer, recovery algorithm and log updater. These proposed algorithms collaboratively work to identify and recover the recovering site from system failure. Quantitative evaluation of our proposed algorithms indicates that our proposed algorithms shall be more effective and efficient than traditional fault recovery techniques.