News

Syncfusion Puts Windows Big Data Platform into Production

• By David Ramel
• February 3, 2016

Syncfusion Inc. moved its Windows-based Big Data platform out of preview and into production with new support for Apache Spark, Apache HBase and Scientific Python. The enterprise software development company claims it's the singular Apache Hadoop-based distribution designed for Windows.

The Syncfusion Big Data Platform is described as "a collection of Big Data tools and frameworks" that features simplified installers and visual development tools designed to shorten the learning curve for getting started with Big Data.

That learning curve for Windows-centric developers was complicated by the extra overhead of messing with Linux virtual machines (VMs) and command-line tools, said Syncfusion, which offers its Windows-native approach running on commodity hardware to eliminate that additional complexity. In fact, the company promises to get developers up and running with a Hadoop cluster within 15 minutes.

With the company's Windows-first focus, developers can run Hadoop jobs and access Apache Hive data with Microsoft's C# programming language, though more "traditional" Big Data languages such as Java, Pig, Hive, Python and Scala are also supported.

As part of the platform, the Syncfusion Big Data Studio gives developers an easy-to-use environment for working with Big Data software such as Pig and Hive and accessing the Hadoop Distributed File System (HDFS).

"The Big Data Studio ships with a local install of the Syncfusion Big Data SDK, which provides a complete working Hadoop distribution right on your laptop," the company said. "No virtual machines are needed, so there is no need to juggle between Linux and Windows. You don't even have to be connected to a cluster to work on Hadoop jobs. You can work with Hadoop on your Windows machine, even when offline, and then deploy to a cluster for production when you are ready."

The Research Triangle Park, N.C., company listed the following enhancements to the platform, in addition to the new support for Spark, HBase and Scientific Python:

• Direct support for managing Oozie jobs.
• Improved integration with Syncfusion's machine learning runtime.
• Support to create and manage pseudo-node Hadoop clusters.
• Enhanced HDFS file browser usability.

Along with on-premises installations, Syncfusion said users can run their own Hadoop clusters on VMs supplied by cloud service providers such as Microsoft Azure and Amazon Web Services (AWS), with customization functionality not found in other cloud-based Hadoop services. Also, the Microsoft partner said, its platform is 100 percent compatible with Azure HDInsight, Microsoft's cloud implementation of Hadoop.

"We are very excited to declare the Syncfusion Big Data Platform a comprehensive, stand-alone Big Data solution for live production environments," said exec Daniel Jebaraj in a statement yesterday. "Its robust feature set and wide-ranging support for tools like Apache Spark, HBase, Pig and Hive make it a vital component for Big Data computing on Windows. Furthermore, the numerous enhancements that have led to this release make it a great option for on-premise and cloud-based cluster deployment."

The platform, which has been in a preview period that started with a September 2014 beta, is no longer totally free, but a free community license is available for those who qualify. The company invited developers to contact it for full pricing details.