Relying on traditional spinning disk-based database platforms isn’t always what is required for modern application deployment. It’s a challenge that VMware is aiming to solve with its new Gemfire 7.0 release.
“With Gemfire, we are focused on enabling the next generation of applications that require an in-memory approach for the data tier, David McJannet, VMware’s director of Cloud and Application Services, told Database Journal.
Gemfire is a standalone in-memory data grid technology that can in some cases replace a traditional database. McJannet said that as performance demands on applications grow, increasing volumes of data are moving to an in-memory approach. VMware acquired the Gemfire in-memory technology in 2010.
McJannet noted that Gemfire can also be used as a distributed caching layer for existing database technology, including VMware’s version of the PostgreSQL database.
“Your data may sleep in a database, but it lives in memory,” McJannet said.
Increasingly though, McJannet said that developers are now using the in-memory grid as the data store of record, rather than just co-existing with a traditional database.
The new Gemfire 7.0 release increases the scale that the in-memory data grid can store. Up to 40 TB of data can now be stored in memory with the new release.
“People want to do analytics around ever larger data sets and make decision in real time,” McJannet said. “The traditional approach of storing that all in a database, doing a query and then waiting for the answer, doesn’t match the requirements for modern online applications.”
Another key new feature in Gemfire 7.0 is native support for JSON. McJannet explained that Gemfire has a rich API that enable lots of features and functionality. That API however required developers to understand the specifics of Gemfire’s API model.
“With JSON, we’re providing what is a very standard document storage type that can now be interacted with by developers,” McJannet said. “So if I’m building a new application, rather than having to learn the specifics of the Gemfire API, I can now just use JSON in conjunction with Gemfire.”
McJannet stressed that the support of JSON greatly simplifies the usability of Gemfire for developers. Moving forward he added that the plan is to continue to broaden the interfaces into the core Gemfire run time platform.
“We’re going to continue to enable a broader set of developers to use familiar constructs to work with Gemfire,” McJannet said. “Document storage is one item, but there is also object storage and key value stores and a whole category of NoSQL types that you’ll see us make richer in Gemfire.”