Two highly efficient algorithms are known for optimally ordering joins while cross products: , which is based on dynamic programming, and Top-Down Partition Search, on memoization. have two severe limitations: handle only (1) simple (binary) join predicates and (2) inner joins. , real queries may contain complex join predicates, involving more than relations, outer joins as well as other non-inner joins. the most efficient known join-ordering algorithm, DPccp, as a starting, first develop a new algorithm, DPhyp, is capable to handle complex join predicates efficiently. do so by modeling the query graph as a (variant of a) hypergraph and then about its subgraphs. , we present a technique to exploit this capability to efficiently handle widest class of non-inner joins dealt with so far. experimental results show that this reformulation of-inner joins as complex predicates can improve optimization by orders of magnitude, compared to known algorithms dealing with complex predicates non-inner joins. again, this gives dynamic programming a distinct advantage over current techniques.
Moerkotte et al. (2008) studied this question.