Review - Concurrency Control Performance Modeling: Alternatives and Implications.

H. V. Jagadish: Review - Concurrency Control Performance Modeling: Alternatives and Implications. ACM SIGMOD Digital Review 2: (2000)

Review

This paper studied, and settled, a burning question of its day, namely the relative benefits of locking versus other more optimisic forms of concurrency control. But the reason I like it so much is not for its impact as for the fact that it is an example worth emulating for how to write a good experimental paper in the database field. It defines the problem clearly, sets up an evaluation system carefully, explains all the parameters involved, and assesses the alternative schemes. Then it goes on to show how sensitive the results are to changing parameter values, and explains in what operating regimes one may prefer which concurrency control alternative. A person who has little knowledge database systems issues can read this paper and walk away with a decent understanding of the performance impacts of concurrency control. Now, if only every performance evaluation paper could be as good ....

References

[1]: Rakesh Agrawal, Michael J. Carey, Miron Livny: Concurrency Control Performance Modeling: Alternatives and Implications. ACM Trans. Database Syst. 12(4): 609-654(1987)