Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/21018#discussion_r181093091 --- Diff: sql/core/src/main/scala/org/apache/spark/sql/execution/CacheManager.scala --- @@ -119,26 +119,60 @@ class CacheManager extends Logging { while (it.hasNext) { val cd = it.next() if (cd.plan.find(_.sameResult(plan)).isDefined) { - cd.cachedRepresentation.cachedColumnBuffers.unpersist(blocking) + cd.cachedRepresentation.clearCache(blocking) it.remove() } } } + /** + * Materialize the cache that refers to the given physical plan. --- End diff -- Why do we need to involve `CacheManager`? Shall we just make the creation of RDD lazy in `InMemoryRelation` and trigger the materialization in `InMemoryTableScanExec`?
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org