sp26-cs544 cs544
Lecture 21: Spark RDDs and Execution Lecture 21: Spark RDDs and Execution
Content locked
The catalog is public. Reading this content requires approved access.
Sign in to readLecture 21: Spark RDDs and Execution
Why replace intermediate HDFS files?
RDDs describe how to obtain data
Lazy computation moves where errors appear
Partitions are units of work
Cache a reusable intermediate result
Driver, executors, and deployment
Connections