Key points are not available for this paper at this time.
Query Optimization is expected to produce good execution plans for complex queries while taking relatively small optimization time. Moreover, it is expected to pick the execution plans with rather limited knowledge of data and without any additional input from the application. We argue that it is worth rethinking this prevalent model of the optimizer. Specifically, we discuss how the optimizer may benefit from leveraging rich usage data and from application input. We conclude with a call to action to further advance query optimization technology.
Surajit Chaudhuri (Mon,) studied this question.