Includes bibliographical references and index.

With the rapid growth of Cloud computing, the size of Cloud data is expanding at a dramatic speed. A huge amount of data is generated and processed by Cloud applications, putting a higher demand on cloud storage. While data reliability should already be a requirement, data in the Cloud needs to be stored in a highly cost-effective manner. This book focuses on the trade-off between data storage cost and data reliability assurance for big data in the Cloud. Throughout the whole Cloud data lifecycle, four major features are presented: first, a novel generic data reliability model for describing data reliability in the Cloud; second, a minimum replication calculation approach for meeting a given data reliability requirement to facilitate data creation; third, a novel cost-effective data reliability assurance mechanism for big data maintenance, which could dramatically reduce the storage space needed in the Cloud; fourth, a cost-effective strategy for facilitating data creation and recovery, which could significantly reduce the energy consumption during data transfer.

1.1. Data reliability in the Cloud -- 1.2. Background of Cloud storage -- 1.2.1. Distinctive features of Cloud storage systems -- 1.2.2. The Cloud data life cycle -- 1.3. Key issues of research -- 1.4. Book overview -- 2.1. Data reliability assurance in hardware -- 2.1.1. Disk -- 2.1.2. Other storage medias -- 2.2. Data reliability assurance in software -- 2.2.1. Replication for data reliability -- 2.2.2. Erasure coding for data reliability -- 2.3. Data transfer for distributed systems -- 2.4. Summary -- 3.1. Motivating example -- 3.1.1. The pulsar searching application process -- 3.1.2. The pulsar searching application data flow -- 3.1.3. Storing pulsar searching data in the Cloud -- 3.2. Problem analysis -- 3.2.1. Two major factors of Cloud storage cost -- 3.2.2. Data storage devices and schemes -- 3.2.3. Cloud network and data transfer activities -- 3.2.4. Research issues -- 3.3. Summary -- 4.1. Properties of the data reliability model -- 4.1.1. Reliability metrics.

4.1.2. Data reliability model type -- 4.1.3. Failure rate pattern of storage devices -- 4.2. Generic data reliability model -- 4.2.1. Data reliability with static disk failure rate -- 4.2.2. Data reliability with variable disk failure rate -- 4.2.3. Generic data reliability model for multi-replicas -- 4.3. Summary -- 5.1. The minimum replication calculation approach -- 5.1.1. Minimum replication calculation formulas -- 5.1.2. Optimization of the minimum replication calculation formulas -- 5.2. Minimum replication benchmark -- 5.3. Evaluation of the minimum replication calculation approach -- 5.4. Summary -- 6.1. Proactive replica checking -- 6.2. Overview of PRCR -- 6.2.1. User interface -- 6.2.2. PRCR node -- 6.3. Working process of PRCR -- 6.4. Optimization algorithms in PRCR -- 6.4.1. Minimum replication algorithm -- 6.4.2. Metadata distribution algorithm -- 6.5. Evaluation of PRCR -- 6.5.1. Performance of PRCR -- 6.5.2. Cost-effectiveness of PRCR.

6.5.3. Summary of the evaluation -- 6.6. Summary -- 7.1. Determining the deadline for data creation and data recovery -- 7.2. Cloud network model -- 7.2.1. Overall network model -- 7.2.2. Pipeline model -- 7.2.3. Pipeline agenda model -- 7.2.4. Overall agenda model -- 7.3. Energy consumption model for Cloud data transfer -- 7.4. Novel cost-effective data transfer strategy LRCDT -- 7.5. Evaluation of LRCDT -- 7.5.1. Parameters of simulation -- 7.5.2. Energy consumption comparison -- 7.5.3. Task completion time comparison -- 7.6. Summary -- 8.1. Summary of this book -- 8.2. Key contributions of this book -- 8.3. Further discussion and future work -- 8.3.1. Further discussions -- 8.3.2. Future work.