Spring for Apache Hadoop 2.0 M5

Spring has happily announced the Spring for Apache Hadoop 2.0 M5 milestone releases, while they are also getting much closer to a release candidate. In the Spring blog there is a good comparison between the new version 2.0 and the 1.0 version. According to it:

1.0 version of Spring for Apache Hadoop uses HDFS and MapReduce with either MapReduce v1 or MapReduce v2 (YARN). The default distribution is Apache Hadoop 1.2.1 with additional features, like Hadoop 2.2.0, Pivotal HD 1.1, Cloudera CDH4 MR1 or MR2 YARN and Hortonworks HDP 1.3.

On the other hand, Spring for Apache Hadoop 2.0 focuses in adding YARN application development support in addition to continue to provide improvements in the HDFS and MapReduce support. The default distribution for the 2.0 releases is Apache Hadoop 2.2.0.

Below you can see the specific artifacts with their respective transitive dependencies in the Spring IO milestone repository:

Spring for Apache Hadoop 2.0 version also offers:

All new YARN features are in the YARN samples, so you can check them out there.

A spring-data-hadoop-store sub-project is also here to provide better support for writing data to HDFS using DataWriter and DataReader implementation supporting formats like text files and SequenceFiles with or without compression. The new sub-project also integrates with the Dataset support from Kite SDK.

For more project specific information please see the project page.

Exit mobile version