Checkpointed RDDs are materialized on disk, while cached RDDs are materialized in memory. When memory is insufficient, cached RDD blocks (1 block per partition) will be evicted in an LRU manner. An evicted RDD block will be spilled to disk if the storage level of the RDD allows, otherwise this block vanishes entirely and must be recomputed from the lineage DAG if it's referenced later.
On Mon, Apr 14, 2014 at 10:20 AM, David Thomas <dt5434...@gmail.com> wrote: > What is the difference between checkpointing and caching an RDD? >