Priority in DBMS resource scheduling
VLDB '89 Proceedings of the 15th international conference on Very large data bases
The PROMPT Real-Time Commit Protocol
IEEE Transactions on Parallel and Distributed Systems
Main Memory Database Systems: An Overview
IEEE Transactions on Knowledge and Data Engineering
Multiclass Query Scheduling in Real-Time Database Systems
IEEE Transactions on Knowledge and Data Engineering
Adaptive Commitment for Real-Time Distributed Transactions
Adaptive Commitment for Real-Time Distributed Transactions
High performance distributed real-time commit protocol
Journal of Systems and Software
Hi-index | 0.01 |
Distributed real time Commit protocols have been an area of extensive research in the past decade and a very less amount of the work [1,2,3,4] has been reported in the literature. However, most of the past research investigate the behavior of these protocols either under read-before-write model or write only (blind write) model. The effect of both types of write models has not been investigated collectively in any previous work. These protocols also require a considerable amount of memory for maintaining temporary objects (data structure) created during execution of transactions. This paper presents a new memory efficient commit protocol (MECP) for distributed real time database systems, in which, write operation is divided into two types: blind write and update. The data items locked by a committing cohort can be lent to a newly arrived higher priority cohort in case of access conflicts same as in [1,2]. The proposed protocol optimizes the memory required for maintaining the information (records of temporary objects) of the lender & borrower cohorts [1] using the data items in conflicting way. We also present the comparative study of our protocol with PROMPT and 2SC commit protocols.