Link Details

Managing large datasets is hard; running computations on large datasets is even harder. Once a dataset has exceeded the capacity of a single filesystem or a single machine, running data processing tasks requires specialist hardware and applications, or, if attempted on a network of commodity machines, it requires a lot of manual work to manage the process: splitting the dataset into manageable chunks, launching jobs, handling failures, and combining job output into a final result. Apache's Hadoop project aims to solve these problems by providing a framework for running large data processing applications on clusters of commodity hardware. Combined with Amazon EC2 for running the application, and Amazon S3 for storing the data, we can run large jobs very economically. This paper describes how to use Amazon Web Services and Hadoop to run an ad hoc analysis on a large collection of web access logs that otherwise would have cost a prohibitive amount in either time or money.