Cloud Data lakes can be used as a replacement for HDFS. In MapReduce job, data is divided into chunks and processed parallelly. Does Data Lakes also split data into chunks?