Products
Cascading
Cascading is a feature rich API for defining and executing complex, scale-free, and fault tolerant data processing work-flows on an Apache Hadoop cluster.
The processing API lets the developer quickly assemble complex distributed processes without having to “think” in MapReduce. And to efficiently schedule them based on their dependencies and other available meta-data. Complex jobs tend to start simple, so Cascading supports simple data processing applications as well, and can effectively absorb additional complexity as applications are developed and matured.
Because Cascading is an alternative API to MapReduce, and not an independent System or Architecture, it easily fits in to organizations development process allowing developers to quickly create unit and integration tests of all aspects of their application.
Cascading allows developers to use an easy-to-visualize, source-pipe-sink paradigm through a Java API instead of writing multiple MapReduce jobs. With Cascading, developers can create new operations or reuse past operations, chain these operations into data processing workflows and save results to an output data set.
Cascading allows developers to chain series of pipes and filters together that read data from data sources and feed it to data sinks.
The Cascading planner converts this assembly of pipes into a cluster-executable processing flow consisting of managed MapReduce jobs.
Finally, this flow is executed on a remote Hadoop cluster in the most efficient manner possible.
Cascading works with both “on-premise” and “cloud” deployed clusters. And is used in production by many companies in Amazon EC2 and Elastic MapReduce. Both the MultiTool and CloudFront LogAnalyzer were developed with Cascading.
Cascading is Open Source and dual licensed under the GPL and OEM/Commercial Licenses. We offer both OEM/Commercial Licenses and Developer Support options for Cascading.
Read more about Cascadings features or thumb through the Cascading 1.0 User Guide. Visit the community site for more.
