Mode of access: World Wide Web.

System requirements: Adobe Acrobat Reader.

Part of: Synthesis digital library of engineering and computer science.

Includes bibliographical references.

part I. Published papers. 1. A survey of B-tree locking techniques -- 1.1. Introduction -- 1.2. Preliminaries -- 1.3. Two forms of B-tree locking -- 1.4. Protecting a B-tree's physical structure -- 1.5. Protecting a B-tree's logical contents -- 1.6. Future directions -- 1.7. Summary and conclusions -- 1.8. References

2. Hierarchical locking in B-tree indexes -- 2.1. Introduction -- 2.2. Related work -- 2.3. Assumptions -- 2.4. Traditional locking hierarchies -- 2.5. Locks on separator keys -- 2.6. Locks on key prefixes -- 2.7. Summary and conclusions -- 2.8. References

3. Concurrent queries and updates in summary views and their indexes -- 3.1. Introduction -- 3.2. Prior work -- 3.3. Multi-version snapshot isolation -- 3.4. Concurrent updates and linear version history -- 3.5. Logging and recovery -- 3.6. Multi-granularity locking -- 3.7. Update and upgrade locks -- 3.8. Insert and delete -- 3.9. Online index operations -- 3.10. Correctness -- 3.11. Performance -- 3.12. Summary and conclusions -- 3.13. References

4. Controlled lock violation / Goetz Graefe, Mark Lillibridge, Harumi Kuno, Joseph Tucek, and Alistair Veitch -- 4.1. Introduction -- 4.2. Related prior work -- 4.3. Controlled lock violation -- 4.4. Distributed transactions -- 4.5. Canned transactions -- 4.6. Performance evaluation -- 4.7. Discussion -- 4.8. Summary and conclusions -- 4.9. Acknowledgements -- 4.10. References

5. Orthogonal key-value locking -- 5.1. Introduction -- 5.2. Prior designs -- 5.3. Orthogonal key-value locking -- 5.4. Case studies -- 5.5. Future opportunities -- 5.6. Conclusions -- 5.7. References

part II. Optimistic concurrency control. 6. Orthogonal key-value validation -- 6.1. Introduction -- 6.2. Related prior work -- 6.3. Orthogonal key-value validation -- 6.4. Case studies -- 6.5. Alternative approaches -- 6.6. Conclusions -- 6.7. References

7. Serializable timestamp validation -- 7.1. Introduction -- 7.2. Related prior work -- 7.3. Phantom protection with timestamp validation -- 7.4. Management of timestamps -- 7.5. Hierarchical timestamps -- 7.6. Conclusions -- 7.7. References

8. Repairing optimistic concurrency control -- 8.1. Introduction -- 8.2. Related prior work -- 8.3. Concurrent validation -- 8.4. Premature publication -- 8.5. Distributed transactions -- 8.6. Conclusions -- 8.7. References

part III. Locking. 9. Avoiding index-navigation deadlocks -- 9.1. Introduction -- 9.2. Related prior work -- 9.3. Recommended locking sequences -- 9.4. Conclusions -- 9.5. References

10. A problem in two-phase commit -- 10.1. References

11. Deferred lock enforcement -- 11.1. Introduction -- 11.2. Related prior work -- 11.3. Deferred lock acquisition -- 11.4. Deferred lock enforcement -- 11.5. Deferred lock enforcement and other techniques -- 11.6. Distributed operations -- 11.7. Summary and conclusions -- 11.8. References

part IV. The end of optimistic concurrency control. 12. The end of optimistic concurrency control -- 12.1. Introduction -- 12.2. Earlier comparisons -- 12.3. Common misconceptions -- 12.4. Locking in optimistic concurrency control -- 12.5. Advantage : locking -- 12.6. Another quick look at optimistic concurrency control -- 12.7. Recommendations -- 12.8. Conclusions -- 12.9. References.

Abstract freely available; full-text restricted to subscribers or individual document purchasers.

Compendex

INSPEC

Google scholar

Google book search

This book contains a number of chapters on transactional database concurrency control. A two-sentence summary of the volume's entire sequence of chapters is this: traditional locking techniques can be improved in multiple dimensions, notably in lock scopes (sizes), lock modes (increment, decrement, and more), lock durations (late acquisition, early release), and lock acquisition sequence (to avoid deadlocks). Even if some of these improvements can be transferred to optimistic concurrency control, notably a fine granularity of concurrency control with serializable transaction isolation including phantom protection, pessimistic concurrency control is categorically superior to optimistic concurrency control, i.e., independent of application, workload, deployment, hardware, and software implementation.

Also available in print.

Title from PDF title page (viewed on June 26, 2019).