A time-cognizant dynamic crash recovery scheme suitable for distributed real-time main memory databases

Authors:
Yingyuan Xiao;Yunsheng Liu;Xiangyang Chen;Xiaofeng Liu
Affiliations:
Department of Computer Science and Engineering, Tianjin University of Technology, Tianjin, P.R. China;Software College, Huazhong University of Science and Technology, Wuhan, Hubei, P.R. China;School of Computer Science and Engineering, Wuhan Institute of Technology, Wuhan, Hubei, P.R. China;Software College, Huazhong University of Science and Technology, Wuhan, Hubei, P.R. China
Venue:
ATC'06 Proceedings of the Third international conference on Autonomic and Trusted Computing
Year:
2006

Citing 4
Cited 0

Rewriting Histories: Recovering from Malicious Transactions

Distributed and Parallel Databases - Security of data and transaction processing
Data dependency based logging for defensive information warfare

SAC '00 Proceedings of the 2000 ACM symposium on Applied computing - Volume 1
Real-Time Database Systems: Architecture and Techniques

Real-Time Database Systems: Architecture and Techniques
Recovery Algorithms for Database Machines with Nonvolatile Main Memory

IWDM '89 Proceedings of the Sixth International Workshop on Database Machines

Quantified Score

Hi-index	0.00

Visualization

Abstract

Rapid and efficient recovery in the event of site crash is very important for distributed real-time main memory database system. In this paper, the recovery correctness criteria of distributed real-time main memory databases are first given. Then, a time-cognizant dynamic crash recovery scheme (TCDCRS) based on log is presented. TCDCRS uses nonvolatile RAM as logging store and integrates the properties of partitioned logging, ephemeral logging and real-time logging in order to reduce the logging cost as possible during the normal running. During restart recovery after site crashes, a dynamic recovery method based on the classification recovery strategy, which supports concurrent of system services and recovery processing, is adopted to decrease the downtime to the most extent. Experiments and evaluations show that TCDCRS has better performances than traditional recovery schemes in two aspects: the missing deadlines ratio of transactions and the time of system denying services after crashes.