Get early access: Please register here to learn more about our Big Data offering and get early access to the software.

Symantec Solution for Enterprise Hadoop is an add-on for customers using Veritas Cluster File System. This easy to install software enables Cluster File System customers run Big Data analytics on their existing infrastructure without investing in a separate cluster just for analytics.

How it works

The solution works within the confines of the modular architecture of Apache Hadoop, using only the public APIs. This allows us to deliver enterprise ready Big Data solution without creating yet another competing Hadoop distribution. This works with Open Source Apache Hadoop and enables customers run standard analytics applications they would run on Apache Hadoop.

Replacing HDFS class: In this solution we provide a software layer that replaces Hadoop Distributed File System (HDFS)  and removes its limitations. In the Hadoop Java class hierarchy, we provide an innovative implementation for FileSystem Java interface using Cluster File System. This implementation is protocol compatible with HDFS and seamlessly supports rest of the Hadoop stack (such as MapReduce). All your MapReduce applications will run on Symantec solution without requiring any changes.

Works on existing CFS installations: The solution works as-is on existing Cluster File System installations. This makes it possible to run analytics on existing data without extracting, transforming and loading data to a separate Hadoop cluster.

Single package install: Adding analytics is as easy as installing a single package and running simple configuration scripts. We include all the JAR files and configuration scripts in  a single package. All you need are Cluster File System, Apache Hadoop distribution and Java runtime environment.

Making Hadoop Highly Available: We provide file system high availability by fixing HDFS’ NameNode issues and add application high availability to MapReduce jobs using Veritas Cluster Server (VCS). Since the Hadoop stack runs on Cluster File System, each node in the cluster can access data simultaneously. Along with high availability for MapReduce, your analytics applications will continue to run as long as there is at least one working node in the cluster.

Easy import / export of data: Any POSIX command like, “cp”, “mv” etc can be used to copy in or copy out data. In addition, this can be done over Network File System (NFS) as well as locally.

This is Big Data customers want, from the infrastructure they’ve got!

Get early access: Please register here to learn more about our Big Data offering and get early access to the software.

Apache Hadoop, Hadoop, HDFS, Avro, Cassandra, Chukwa, HBase, Hive, Mahout, Pig, Zookeeper are trademarks of the Apache Software Foundation.