Open source Hadoop vendor MapR Technologies Inc. is expanding the functionality of its Big Data platform with the release of MapR 4.1, with the headline feature being new table replication features designed to extend access to “big and fast” data. The release comes as the company unwraps a trio of Quick Start Solutions it says will help customers to jump start their Hadoop deployments.

MapR has long set itself apart from competitors Cloudera Inc. and Hortonworks Inc. with its ability to make the batch-orientated Hadoop database act as if it’s a real-time database, and version 4.1 of its flagship product takes that concept further, explained Jack Norris, the company’s Chief Marketing Officer. With the addition of table replication, it becomes possible for multiple instances to be updated in different locations, with all the changes synchronized across them. The feature also reduces the risk of users suffering any data loss due to widespread system failures, as it helps to enable real-time disaster recovery, while the ability to store and process data close to users and devices delivers enhanced application deployment.

“If you’ve got data and a replicated database, you can send changes across the wire to effect changes at both ends,” said Norris. “We were the only ones who had WAN replication but we now can have two primary databases at two locations. Two separate stores can capture transactions and update each other’s databases at the same time.”

The latest version of MapR Hadoop also introduces a new C language API for MapR-DB which lets developers write real-time applications in the language, plus a new MapR POSIX Client that delivers critical speed and security advantages to applications running on edge nodes. The latter addition should prove to be especially useful for organizations dealing with larger data sets, Norris claimed, because it expands and enables new ways for that data to be digested.

“The previous option was to load the data as a Hadoop app and come in through the distributed file system or you could mount it as a gateway,” said Norris. “Now you have transport through the MapR clients for better speed and security.”

Alongside a more capable Hadoop platform, MapR believes it can help to deliver a faster return on investment for its customers. The company already makes the rather impressive claim that most of its users see a payback within 12 months or less, but it believes it can deliver results even sooner. To that end, it’s rolling out a series of new “Quick Start Solutions” packages designed to help customers get off the ground in the fastest possible time.

Norris told SiliconANGLE the solutions, which combine deployment best practices with recommended tools and services in the Hadoop ecosystem to minimize the learning curve for customers, are likely to just the first of a family of such quick-start offerings. For starters there are three to choose from, each one tailored for a specific solution area.

Options include a new Data Warehouse Optimization and Analytics solution that ensures all transactions are performed on Hadoop, thereby freeing up storage space to reduce overall system costs. Next is its Security Log Analytics solution that helps to detect advanced and unknown threats using deep analytics and anomalous behavior prediction techniques. Finally, MapR is touting its Recommendation Engine solution, which it says combines past transactions and other customer attributes to deliver real-time offers for products and services, thereby boosting customer revenues and loyalty.

“We look at this as an underpinning for many applications. Hadoop today is used to reduce costs and mitigate risks and each of these packages is aligned to a different aspect of those use cases,” Norris said.

MapR version 4.1 is available from today, and the new Quick Start Solutions are also available now, with pricing starting at $30,000.

Image credit: Alan9187 via Pixabay.com