VMware is introducing its Spring Hadoop software, which is intended to make it easier for Java developers using the Spring Framework to leverage Apache Hadoop data processing capabilities.
Developers can perform MapReduce queries in Hadoop from Spring, then have triggered event results based on Hadoop, said Adam Fitzgerald, VMware director of developer relations. Also, developers can build complex workloads that interact with Hadoop either as individual MapReduce requests or as data-streaming results.
Hadoop is Apache's open source platform for scalable, distributed computing, while Hadoop MapReduce is a programming model and framework for processing large sets of data. Spring Hadoop will be available on VMware's springsource.org website and is being released under an Apache open source licence.
"Spring Hadoop was created to make it more straightforward for enterprise Java developers to use Apache Hadoop," Fitzgerald said. With the integration, VMWare has taken Spring's dependency injection mechanism for linking related objects and applied it to Hadoop. This saves developers time and increases productivity, testability and portability, Fitzgerald said.
Spring Hadoop enables execution of MapReduce, Streaming, Hive, Pig and Cascading jobs via the Spring container. Hadoop Distributed File System data access is enabled through JVM scripting languages, such as Groovy and jRuby. Also, declarative and programmatic support is offered for Hadoop tools, including FsShell and DistCp.