Hadoop has been updated with improvements including support for ARM architectures, Java 11 runtime support, and S3A enhancements.

Hadoop is a framework that can be used to process large data sets across clusters of computers using simple programming models. It includes the HDFS filing system, the YARN framework for job scheduling and cluster resource management and high availability for the HDFS filing system, and MapReduce, a framework for parallel processing of large data sets across a Hadoop cluster.

Apache Hadoop 3.3 is the first release of Hadoop this year, and is the first release to support ARM architectures. Other improvements include a variety to the S3A client, which offers high-performance IO against Amazon S3 object store and compatible implementations. The client has been improved with the addition of Delegation Token support, better handling of 404 caching, and improved S3guard performance. S3Guard is a feature for the S3A client of the S3 object store, which can use a consistent database as the store of metadata about objects in an S3 bucket.

ABFS has also been improved in this release. The ABFS connector provides support for the Azure Data Lake Storage Gen2 storage layer, and in this release has been tuned for better performance, and troubleshooting options have been improved.

A new format that's been added is Tencent Cloud COS File System. Tencent Cloud is one of the top two cloud vendors in the Chinese market, and the object store COS is widely used among China’s cloud users. This release adds a COSN filesytem to support Tencent cloud COS natively in Hadoop.

Other improvements include stabilization of the HDFS RBF, so the router now supports security; and Protobuf has been upgraded to version 3.7.1.

More Information

Apache Hadoop Site 

Related Articles

Hadoop 2.9 Adds Resource Estimator

Hadoop Adds In-Memory Caching

Hadoop SQL Query Engine Launched

Hadoop 2 Introduces YARN

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info