"If the Spark Driver fails, it will be replaced by Worker node" Is it a correct statement ? Does this depend on the Cluster manager ?
