But analytics remains a major draw for new users of in-memory data grids, said Jon Webster, vice president of business development at GridGain. “We take an in-memory data grid and we also have a processing data grid. We give you a cohesive API. That’s where you get the processing plus storage. If you look at some of the older caching technologies, you’re still dealing with the pattern of ‘The data is stored somewhere; even if it’s in Coherence, take it out of there and put it back into an application server and process it somewhere else.’ The real problem is that moving that data is exceptionally expensive,” he said.
To that end, GridGain uses that cohesive API to allow developers to analyze data stored in their in-memory data grid without having to move it to another platform. When dealing with fresh terabytes of data every day, that can save developers a great deal of time.
You host it for me
The popularity of in-memory data grids has even spawned a new cloud-based service. Ofer Bengal, cofounder and CEO of Garantia Data, said that his company has taken the popular NoSQL data store Redis and turned it into a scalable data grid platform hosted within the company’s own cloud service.
While Redis is an open-source key-value store, Garantia Data turns the software into an automatically scalable in-memory data grid for applications that need terabytes of quickly accessible information.
“When it comes to replicating Redis onto disk, or into persistent storage, you’ll normally see performance degradation,” said Bengal. “We have overcome this problem, and we offer real-time applications to persistence storage without any degradation of performance. We also offer instant fail-over.”
And, at the end of the in-memory day, the usefulness always comes back to the application being tied into the grid. “More and more, we see in-memory data grid technology being embedded into other software products,” said Gartner’s Pezzini. “Enterprise servers, VPN tools, packaged applications. Of course this is going to help a lot with the establishment of this market. We are seeing very interesting dynamics at play here in this market.”
And while the market is growing quickly, Gartner does not yet have a magic quadrant for in-memory data grids. “I think TIBCO could be the rising star here, and could become an important player in this market,” said Pezzini. “Software AG with Terracotta will be growing primarily because of their software marketing organization. [Software AG] is going to leverage Terracotta inside many webMethods products, and a bunch of new offerings they have there in the making around Big Data.”