In order to allow users to easily take advantage of distributed cloud resources, we built Resilin, a system for executing MapReduce computations. Resilin implements the Amazon Elastic MapReduce web service API and federates resources from scientific and private clouds. Resilin takes care of provisioning, configuring and managing cloud-based Hadoop execution platforms, potentially using multiple clouds. Resilin has been evaluated against the Amazon Elastic MapReduce system and shows similar performance.
More details: Thesis