Syncfusion Ships Production Release for its Big Data Platform for Windows
Syncfusion Inc. is shipping its first production release for its Hadoop distribution -- designed expressly for Windows. The rollout of Syncfusion Big Data Platform come more than a year after its beta release of big data tools and frameworks to support Apache Spark, Apache HBase, and Scientific Python -- along with many other Hadoop must haves.
Syncfusion Inc. is shipping its first production release for its Hadoop distribution -- designed expressly for Windows.
The rollout of Syncfusion Big Data Platform come more than a year after its beta release of big data tools and frameworks to support Apache Spark, Apache HBase, and Scientific Python -- along with many other Hadoop must haves.
“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,” said Daniel Jebaraj, vice president of Syncfusion in a statement. “We are very excited to declare the Syncfusion Big Data Platform a comprehensive, standalone big data solution for live production environments.”
Since its initial beta release in September 2014, Syncfusion designed its Big Data Platform to deliver a complete environment for creating full clusters on commodity hardware and running Hadoop jobs on them in a scalable manner -- all within the familiar Windows environment.
This first production release of Syncfusion’s Big Data Platform brings together many top-of-mind big data features to help adopters support on-premises and cloud-based cluster deployment, Jebaraj added. Among them:
Easy-to-adopt developer tools
With easy-to-use installers and visual development tools, Syncfusion aims to reduce learning curve for working with data. Further, developers can write Hadoop jobs using their favorite language, including Java, Hive, Python, and Scala, C #or other .NET platform language.
Enterprise-grade Hadoop cluster
A complete production environment that can run Hadoop jobs in a scalable manner on a full cluster. An included ‘Cluster Manager’ application makes provisioning easy, allowing users to manage and monitor multiple-node Hadoop clusters on Windows.
Easy-to-create production Hadoop clusters
Using Syncfusion’s Hadoop distribution, users can create clusters in minutes using commodity machines running the most recent versions of Windows (Windows 7/Windows Server 2008 and later).
Management and monitoring for multiple clusters
Syncfusion’s cluster management system comes with built-in support for managing multiple clusters. Easily manage and monitor multiple clusters from a web browser.
Easily monitor job status
A bundled monitoring system provides information on the general health of clusters. It also provides detailed metrics on nodes that are part of the cluster and jobs running on the cluster.
Integrated HBase support
Integrated support for HBase, quickly becoming a top NoSQL store, lets users store and process massive amounts of data in a horizontally scalable manner, and achieve high speed reads and writes.
No requirement for manual configuration
When creating a cluster, manual configuration is not required. The cluster configuration and management system takes care of everything
Automate agent installation with Windows PowerShell/Active Directory
While the agent can be manually installed for smaller clusters, Syncfusion provide scripts that can automate the installation of agents to a large number of nodes in a completely unattended manner.
Ready for YARN applications, users can run any YARN-compatible application system – breaking any reliance on MapReduce.
- Syncfusion’s Big Data Platform also offers
- Direct support for managing Apache Oozie (workflow scheduler for Hadoop) jobs.
- Improved integration with Syncfusion’s machine learning runtime.
- Support to create and manage pseudo-node Hadoop clusters.
- Enhanced HDFS file browser usability.
Darren West, a content producer at Syncfusion highlighted the impact of many of Syncfusion Big Data Platform features in a recent blog post.
Headlining the release is Spark and HBase integration across all components of the platform, including the studio, cluster manager, C# libraries, and SDK. This may significantly improve performance for certain applications, and may provide vastly better data storage options for developers. HBase support allows Big Data Platform users to store and process massive amounts of data.
Scientific Python support adds new capabilities, including Python package management. In providing built-in support for open source solutions, Syncfusion remains committed to meeting the practical demands of our customers.
In addition . . . the production release includes many improvements and updates over the beta version. Improvements to the Syncfusion Big Data Studio include cluster communication behavior when working with a remote agent, improved console editor usability, and updated homepage design with backstage view to add and remove clusters. It has also received a number of usability enhancements to improve the HDFS file browser, Hive database and HBase table tree view, and the local service manager. The Cluster Manager application now supports creating and managing pseudo-node Hadoop clusters, starting and stopping all thrift services and Oozie servers, and submitting and monitoring Oozie jobs. It also has enhanced Hadoop support. Finally, the C# libraries and SDK see a variety of support updates for Spark, Hive, HBase, and other solutions.
A free 30-day trial is available from Syncfusion. Users can start with either a 5- or 10-node cluster, or scale as they need to, Jebaraj added.