Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/17546#discussion_r110570339 --- Diff: sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/CostBasedJoinReorder.scala --- @@ -327,3 +345,104 @@ object JoinReorderDP extends PredicateHelper with Logging { case class Cost(card: BigInt, size: BigInt) { def +(other: Cost): Cost = Cost(this.card + other.card, this.size + other.size) } + +/** + * Implements optional filters to reduce the search space for join enumeration. + * + * 1) Star-join filters: Plan star-joins together since they are assumed + * to have an optimal execution based on their RI relationship. + * 2) Cartesian products: Defer their planning later in the graph to avoid + * large intermediate results (expanding joins, in general). + * 3) Composite inners: Don't generate "bushy tree" plans to avoid materializing + * intermediate results. + * + * Filters (2) and (3) are not implemented. + */ +case class JoinReorderDPFilters(conf: SQLConf) extends PredicateHelper { + /** + * Builds join graph information to be used by the filtering strategies. + * Currently, it builds the sets of star/non-star joins. + * It can be extended with the sets of connected/unconnected joins, which + * can be used to filter Cartesian products. + */ + def buildJoinGraphInfo( + items: Seq[LogicalPlan], + conditions: Set[Expression], + planIndex: Seq[(LogicalPlan, Int)]): Option[JoinGraphInfo] = { + + // Compute the tables in a star-schema relationship. + val starJoin = StarSchemaDetection(conf).findStarJoins(items, conditions.toSeq) + val nonStarJoin = items.filterNot(starJoin.contains(_)) + + if (starJoin.nonEmpty && nonStarJoin.nonEmpty) { + val (starInt, nonStarInt) = planIndex.collect { + case (p, i) if starJoin.contains(p) => + (Some(i), None) + case (p, i) if nonStarJoin.contains(p) => + (None, Some(i)) + case _ => + (None, None) + }.unzip + Some(JoinGraphInfo(starInt.flatten.toSet, nonStarInt.flatten.toSet)) + } else { + // Nothing interesting to return. + None + } + } + + /** + * Applies star-join filter. + * + * Given the outer/inner and the star/non-star sets, + * the following plan combinations are allowed: + * 1) (outer U inner) is a subset of star-join + * 2) star-join is a subset of (outer U inner) + * 3) (outer U inner) is a subset of non star-join + * + * It assumes the sets are disjoint. + * + * Example query graph: + * + * t1 d1 - t2 - t3 + * \ / + * f1 + * | + * d2 + * + * star: {d1, f1, d2} + * non-star: {t2, t1, t3} --- End diff -- can we also have an example about `outer` and `inner`?
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org