ABSTRACT
Set manipulation consists of the actions insert, delete, and member on keys. We propose a concurrent set manipulation algorithm that uses no locking at all and requires no aborts, relying instead on atomic read-modify-write operations on single (data) locations. The algorithm satisfies order-preserving serializability through conditions that are strictly looser than existing algorithms
- BG81.- P A Bernstem and N Goodman, "Concurrency Control m Dtstrlbuted Database Systems," Compuung Surveys 13, " (June 1981), pp 185-221]] Google ScholarDigital Library
- BS77.- R Bayer and M Schkolmck, "Concurrency of operauons on B-trees," Acta Informattca, vol 9, pp 1- 21, 1977]]Google ScholarDigital Library
- El80.- C S Elhs, "Concurrent Search and Inserts in 2-3 trees," Acta Informatlca, vol 14, no 1, pp 63-86, 1980]]Google ScholarDigital Library
- El82.- C S Elhs, "Extendible Hashing for Concurrent Operatmns and D~strlbuted Data," ACM-SIGMOD, pp 106-115, 1982]] Google ScholarDigital Library
- El85.- C S Elhs, "Concurrency control and hnear hashing," fourth annual ACM SIGACF/SIGMOD symposium on Prmclples of Database Systems, Portland, Oregon, pp 1- 7, 1985]]Google Scholar
- FC84.- R Ford and J Calhoun, "Concurrency Control Mechamsms and the Senahzablhty of Concurrent Tree Algorithms," Proceeclmgs of the ACM Symposmm on the Prmclples of Database Systems, 1984]] Google ScholarDigital Library
- GGKMRS83.- A Gottheb, R Gnshman, C K Kruskal, K P McAuhffe, L Rudolph, and M Snlr, The NYU Ultracomputer - Destgmng a MIMD, Shared memory Parallel Computer IEEE Trans on Computers C-32 (1983), 175-189]]Google Scholar
- GS85.- N Goodman and D Shasha, "Semantically-based concurrency control for search structures," Proc of the fourth annual ACM SIGACT/SIGMOD symposmm on prmctples of database systems, March 1985, pp 8-19]] Google ScholarDigital Library
- HY86.- M Hsu and W P Yang, "Concurrent Operattons m Extendtble Hashing" Proceedsngs of the Twelfth Very Large Data Bases Conference, Kyoto, Japan, August 25-28, 1986 pp 241-247]] Google ScholarDigital Library
- KL80.- H T Kung and P L Lehman, "Concurrent Mampulatton of Binary Search Trees," ACM Trans Database Syst, vol 5, no 3, pp 339 -353, 1980]] Google ScholarDigital Library
- KP83.- H T Kung and C H Papadtmstrtou, "An Opttmahty Theory of Concurrency Control for Databases," Acta Informatlca, vol 19, no 1 (1983), pp 1-11]]Google ScholarDigital Library
- KS83.- Z Kedem and A Stlberschatz, "Locking Protocols From Exclustve to Shared Locks," Journal of the ACM, Vol 30, No 4, Oct 1983, pp 787-804]] Google ScholarDigital Library
- KW82.- Y S Kwong and D Wood, "Method for concurrency m B-trees," IEEE Transacttons on Software Engineering, vol SE-8, no 3, pp 211-223, 1982]]Google ScholarDigital Library
- LS86.- V Lanm and D Shasha, "A symmetrtc concurrent B tree algorithm," Proc of the Fall Joint Computer Conference of 1986, October 1986, pp 380-389]] Google ScholarDigital Library
- LY81.- P L Lehman and S B Yao "Efficient locking for concurrent operations on B-trees," ACM Trans on Database Systems, vol 6, no 4, December 1981, pp 650-670]] Google ScholarDigital Library
- ML82.- U Manber and R E Ladner, "Concurrency control m a dynamic search structure," ACM Proceedings on Database Systems, pp 268-282, 1982]] Google ScholarDigital Library
- MR85.- Y Mond and Y Raz, "Concurrency control in B+ trees using Preparatory Operations," llth International Conference on Very Large Data Bases, August 1985, pp 331-334]]Google Scholar
- Pa79.- C H Papadlmltrlou, "Sertahzablhty of Concurrent Database Updates," JACM, vo 26, no 4, pp 631-653 October, 1979]] Google ScholarDigital Library
- Sag85.- Y Saglv, "Concurrent Operations on 8*-trees with Overtaking," Proc of the fourth annual ACM SIGACT/SIGMOD symposium on principles of database systems, March 1985, pp 28-37]] Google ScholarDigital Library
- Sam76.B Samadi, "B-trees in a System with Muluple Users," Informanon Processing Letters, Voi 5, pp 107-112, October 1976]]Google ScholarCross Ref
- Sch78.G Schlageter, "Process Synchromzatton m Database Systems," ACM Transactions on Database Systems 3,3 (Sept 1978) p 248]] Google ScholarDigital Library
- Sh84.- D Shasha, "Concurrent Algorithms for Search Structures," Ph D Thesis, Harvard Umverslty, TR-12- 84, June, 1984]]Google Scholar
- SLS87.D Shasha, V Lamn, and J Schmtdt, "An Analytical Model for the Performance of Concurrent B Tree Algorithms," New York Umverstty, UltraComputer Note #124, June 1987]]Google Scholar
Recommendations
General and efficient locking without blocking
MSPC '08: Proceedings of the 2008 ACM SIGPLAN workshop on Memory systems performance and correctness: held in conjunction with the Thirteenth International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS '08)Standard concurrency control mechanisms offer a trade-off: Transactional memory approaches maximize concurrency, but suffer high overheads and cost for retrying in the case of actual contention. Locking offers lower overheads, but typically reduces ...
Pre-analysis locking: a safe and deadlock free locking policy
VLDB '85: Proceedings of the 11th international conference on Very Large Data Bases - Volume 11A safe and deadlock free lock policy is introduced, called pre-analysis locking. Pre-analysis locking is based on an efficient geometric algorithm which inserts lock and unlock operations into the transactions. Pre-analysis locking is the first safe and ...
Locking Timestamps versus Locking Objects
PODC '18: Proceedings of the 2018 ACM Symposium on Principles of Distributed ComputingWe present multiversion timestamp locking (MVTL), a new genre of multiversion concurrency control algorithms for serializable transactions. The key idea behind MVTL is simple: lock individual timestamps instead of locking objects. After presenting a ...
Comments