An Application-Based Adaptive Replica Consistency for Cloud Storage

The intrinsic characteristic heterogeneous of cloud applications makes their consistency requirements different. Furthermore, the consistency requirement of certain application changes continuously at runtime, so a fixed consistency strategy is not enough. An application-based adaptive mechanism of replica consistency is proposed in this paper. We divide the consistency of applications into four categories according to their read frequencies and update frequencies, and then design corresponding consistency strategies. Applications select the most suitable strategy automatically at runtime to achieve a dynamic balance between consistency, availability, and performance. Evaluation results show that the proposed mechanism decreases the amount of operations significantly while guaranteeing the application’s consistency requirement.

[1]  Amin Vahdat,et al.  The costs and limits of availability for replicated services , 2001, TOCS.

[2]  Yahya Slimani,et al.  Consistency Management for Data Grid in OptorSim Simulator , 2007, 2007 International Conference on Multimedia and Ubiquitous Engineering (MUE'07).

[3]  Gustavo Alonso,et al.  Consistency Rationing in the Cloud: Pay only when it matters , 2009, Proc. VLDB Endow..

[4]  Amin Vahdat,et al.  The costs and limits of availability for replicated services , 2006 .

[5]  Ruay-Shiung Chang,et al.  Adaptable Replica Consistency Service for Data Grids , 2006, Third International Conference on Information Technology: New Generations (ITNG'06).

[6]  Françoise André,et al.  Dynamic data replication and consistency in mobile environments , 2005, DSM '05.

[7]  John B. Carter,et al.  Flexible Consistency for Wide Area Peer Replication , 2005, 25th IEEE International Conference on Distributed Computing Systems (ICDCS'05).

[8]  Yahya Slimani,et al.  A Hybrid Approach for Consistency Management in Large Scale Systems , 2006, International conference on Networking and Services (ICNS'06).

[9]  Amin Vahdat,et al.  Design and evaluation of a conit-based continuous consistency model for replicated services , 2002, TOCS.

[10]  Zhou Xu,et al.  A dynamic distributed replica management mechanism based on accessing frequency detecting , 2004, OPSR.

[11]  Tim Kraska,et al.  Building a database on S3 , 2008, SIGMOD Conference.