Communications of the ACM
Technical Perspective: Broadening and Deepening Query Optimization Yet Still Making Progress
Query optimization is a fundamental problem in data management. Simply put, most database query languages are declarative rather than imperative—that is, they specify properties the answer should satisfy, rather than give an algorithm to compute the answer. The best known and most widely used database query language—SQL—is a prime example of a language for which optimization is essential.
By "essential," I mean that database optimization is not a matter of shaving 10% or even a factor of 2x from a query's execution time. In database query evaluation, the difference between a good plan and a bad or even average plan can be multiple orders of magnitude—so successful query optimization makes the difference between a plan that runs quickly and one that never finishes at all. Accordingly, since the seminal papers in the 1970s, query optimization has received and continues to receive a great deal of attention from both the industrial and research database communities.
No entries found