Adaptive intersection-point plan-space exploration

Develop an adaptive intersection-point-based plan-space exploration strategy for cost model parameter tuning that uses explicit linear cost functions to reduce the number of query optimizer calls.

Background

The paper’s binary space partitioning (BSP) method is complete but may invoke the query optimizer substantially more times than necessary. The authors describe an alternative strategy that would identify intersection points between linear plan-cost functions and partition the cost-model-parameter space accordingly. Implementing this strategy requires exposing explicit linear cost expressions from the query optimizer, which would entail nontrivial optimizer modifications.

References

One problem of implementing this intersection-point based idea is that we need the explicit expressions of the linear cost functions, which requires nontrivial changes to the query optimizer. We therefore leave the implementation of this intriguing idea to reduce the number of query optimizer calls for future work.

— Query Performance Tuning with Optimal Exploration of Optimizer Cost Model Parameter Space  (2610.02607 - Wu et al., 2 Oct 2026) in Section 2.3.2, “Further Optimization”