Optimistic voting for managing replicated data
Journal of Computer Science and Technology
Reducing transaction abort rate of epidemic algorithm in replicated databases
ICCS'05 Proceedings of the 5th international conference on Computational Science - Volume Part I
| Hi-index | 0.00 |
In this paper, a new technique called transactions sharing quorum technique (TSQT) is proposed to manage replicated data in a distributed database system. This technique provides high availability for read and writes operations by taking into account a number of transactions sharing on data copies. We start by generalizing this technique subject to the transactions sharing data item in common in order to evaluate a majority quorum for both read and write operations. The majority quorum under TSQT is based on two types, in which for type 1, transactions are accessing the same data item at the same time at any point of time and for type II, it is based on specific time interval. Compared to other protocols, both types under TSQT require significantly lower communication costs for an operation, while providing higher data availability.