I've seen dramatic performance improvements from using cached plans. The numbers are like "20ms to plan vs 1ms to execute" for an often used OLTP query. Query text is involved (~5-10KiB).
but currently we have not any tool how to check the quality of plan for new set of parameters. If plan is ok for one value parameters, then can be pretty bad for following parameters.
Albe's proposal can be good enough for 2/3 cases and it doesn't block any other enhancing. There is still 1/3 of queries - too complex (slow planning) too dynamic plan (the generic plan doesn't work).
