How do we handle (a small number of) worker failures when Spark tasks are executed

A. Task Retry
B. Task Duplication
C. Task Repartitioning
D. Task Exclusion