Differentiated Replication Strategy in Data Centers

Cloud computing has attracted a great deal of attention in both academia and industry. We envision the provisioning of differentiated services as being one of the key components to the success of cloud computing. Unfortunately, this issue has not been fully addressed in the past. Realizing that different users might have different requirements regarding availability, reliability, durability, response time and so on, we conjecture that providing flexible replication mechanism is the right approach to service differentiation. In this paper, we propose Differentiated Replication (DiR), which allows users to choose different replication strategies by considering both the user requirements and system capability. We implemented a system that offers four differentiated storage services with DiR. The experimental results show that this service actually provides different availabilities and execution times for different service types with the same request traces, failure traces, and workload. In addition, we also show that in comparison to the regular uniform replication, DiR can further improve resource utilization, which will in turn provide a better user experience with regards to cloud computing.

[1]  Edith Cohen,et al.  Search and replication in unstructured peer-to-peer networks , 2002 .

[2]  Randy H. Katz,et al.  Above the Clouds: A Berkeley View of Cloud Computing , 2009 .

[3]  Edith Cohen,et al.  Replication strategies in unstructured peer-to-peer networks , 2002, SIGCOMM.

[4]  Ming Zhong,et al.  Replication degree customization for high availability , 2008, Eurosys '08.

[5]  Ian T. Foster,et al.  The Globus Replica Location Service: Design and Experience , 2009, IEEE Transactions on Parallel and Distributed Systems.

[6]  Randal E. Bryant,et al.  Data-Intensive Supercomputing: The case for DISC , 2007 .

[7]  Richard P. Martin,et al.  Autonomous replication for high availability in unstructured P2P systems , 2003, 22nd International Symposium on Reliable Distributed Systems, 2003. Proceedings..

[8]  Antony I. T. Rowstron,et al.  Pastry: Scalable, Decentralized Object Location, and Routing for Large-Scale Peer-to-Peer Systems , 2001, Middleware.

[9]  Roger Wattenhofer,et al.  Optimizing file availability in a secure serverless distributed file system , 2001, Proceedings 20th IEEE Symposium on Reliable Distributed Systems.

[10]  Ann L. Chervenak,et al.  A data placement service for petascale applications , 2007, International Parallel Data Systems Workshop.

[11]  Randy H. Katz,et al.  A view of cloud computing , 2010, CACM.

[12]  David R. Karger,et al.  Chord: A scalable peer-to-peer lookup service for internet applications , 2001, SIGCOMM '01.

[13]  Robert Morris,et al.  Chord: A scalable peer-to-peer lookup service for internet applications , 2001, SIGCOMM 2001.

[14]  Geoffrey H. Kuenning,et al.  Introducing permuted states for analyzing conflict rates in optimistic replication , 2005, SIGMETRICS '05.

[15]  Ralf Steinmetz,et al.  The effectiveness of realistic replication strategies on quality of availability for peer-to-peer systems , 2003, Proceedings Third International Conference on Peer-to-Peer Computing (P2P2003).

[16]  Andreas Haeberlen,et al.  Efficient Replica Maintenance for Distributed Storage Systems , 2006, NSDI.

[17]  Shyamala Doraimani,et al.  Filecules: A New Granularity for Resource Management in Grids , 2007 .

[18]  Peter Druschel,et al.  Pastry: Scalable, distributed object location and routing for large-scale peer-to- , 2001 .

[19]  Marvin Theimer,et al.  Feasibility of a serverless distributed file system deployed on an existing set of desktop PCs , 2000, SIGMETRICS '00.

[20]  Edith Cohen,et al.  Search and replication in unstructured peer-to-peer networks , 2002, ICS '02.

[21]  Stefan Savage,et al.  Total Recall: System Support for Automated Availability Management , 2004, NSDI.

[22]  Wei Chen,et al.  On the Impact of Replica Placement to the Reliability of Distributed Brick Storage Systems , 2005, 25th IEEE International Conference on Distributed Computing Systems (ICDCS'05).

[23]  Mark Handley,et al.  A scalable content-addressable network , 2001, SIGCOMM '01.