Hadoop allows you to execute every task in parallel, and the only concern you should have is how many machines are available, and what is the optimal number of machines to use when performing a Map/Reduce job. Instead, the problem is more a question of how you can import and export data from the Hadoop cluster. Accumulo solves this problem by giving applications BigTable access to the Hadoop filesystem.
Accumulo operates over the Hadoop Distributed File System (HDFS). Accumulo supports efficient storage and retrieval of structured data, including queries for ranges, and provides support for using Accumulo tables as the input and output for Map/Reduce jobs.