News

VMware Unveils GemFire 7.0 In-Memory Data Platform

• By John K. Waters
• November 6, 2012

VMware has launched the latest version of its in-memory data platform, vFabric GemFire 7.0, which offers new features aimed at making it easier for Java developers to build apps that process massive volumes of data and scale within the cloud.

This release of GemFire comes with amped-up caching and high-speed parallel WAN processing, which supports low-latency apps that can scale to handle terabytes of data and hundreds of thousands of users. Version 7.0 also adds a new HTML5-based dashboard that provides a single console for monitoring performance across distributed vFabric environments. There's also a simplified command line interface and new "DataBrowser" functionality.

"GemFire has been focused on this In-Memory Data Grid (IMDG) technology for the better part of a decade," said David McJannet director of cloud and application services in VMware's Application Platform group. "And we've seen a huge increase in the amount of interest in this approach to computing, moving away from using a traditional database to build applications. Increasingly people want to build applications using an in-memory construct."

McJannet's group originated with the company's 2009 acquisition of SpringSource, the chief commercial sponsor of the open-source, Java-based Spring Framework project. What became the company's SpringSource division acquired data management vendor GemStone in 2010, and set out almost immediately to use the GemFire enterprise data fabric to provide Java devs using Spring with the infrastructure they needed for emerging cloud-centric applications. A data fabric is a type of grid-based middle-tier data platform. GemStone's flagship GemFire object-oriented fabric provides an in-memory distributed data management platform that pools memory, CPU, network, and local disk across multiple processes to manage application objects and behavior.

"There are a whole bunch of industry trends circling around the need to put more data into memory," McJannet told ADTmag. "That's really the history of GemFire from its inception. As we've seen real shifts in, well, hardware as much as anything, it's become much simpler to use these kinds of technologies. Even five years ago, to store a terabyte of data memory so that I could build a rapidly responding application would have been extremely expensive, and so available only to very large organizations. As memory has become less expensive today, you see more and more organizations wanting to pursue this approach."

GemFire 7.0 is integrated with the Spring framework and the Spring Data GemFire project. That integration includes documentation and code samples designed to make it easier to develop and support scalable, transactional enterprise apps. The company says that apps using Memcached can transition smoothly to vFabric GemFire "with minimal to zero developer impact, since vFabric GemFire 7.0 supports the Memcached client protocol."

This release also provides native support for JavaScript Object Notation (JSON) documents. With the ability to query between multiple JSON docs and across JSON docs and key/value objects, McJannet said, GemFire can be effectively used as a production-ready backend document store for REST and mobile applications.

"If there's a key improvement in this release, I'd have to say its simplification for developers," McJannet said. "We've significantly enhanced the Spring GemFire project, which means if I'm a Spring developer I have a very simple set of methods to put and get objects into GemFire. And then there's the support for a JSON document type within GemFire."

VMware also offers a "sister product" called SQLfire, which provides a SQL interface to this in-memory data grid. GemFire provides the more traditional, rich-client API for putting and getting objects into the data grid, and now for putting and getting JSON documents into the grid.

VMware's vFabric GemFire 7.0 is available now. A trial download can be found here.