MapR Extends Platform for AI, More Security Tools
MapR Technologies has added a number of new features to the MapR Data Platform, including the addition of direct support for artificial intelligence (AI) applications.
The new capabilities the platform now offers help to further extend the number of cloud and server structures it can accommodate on a single platform with persistent storage capabilities. An organization’s AI application developers, for example, can share and access the same data platform within their entire organization across multi-protocol and different on-premises, edge and cloud deployments.
This means “the data scientists, admins, data analysts and the entire DevOps team can have access to the same foundational meanings of the data,” says Anoop Dawar, MapR’s senior vice president.
“Whether you like it or not, we are in a multi-platform and -cloud deployment world,” Dawar says. “What has changed is how it is necessary for all teams in an organization to be able to continually work on the same data set, which MapR allows you to do.”
More specifically, MapR now offers object tiering, ingestion and erasure coding data storage and retention capabilities, as well as more robust security tools.
The new “secure by default” feature, for example, automatically enables a homogenous security layer upon installation for what MapR describes as “wire-level encryption.” It offers authentication across the entire platform for a wide range of cloud and on-premises server deployments upon installation.
“MapR does all of the work up front, including the complex configurations that are required to secure the system automatically when you install it,” Dawar said. “For developers, this means they can get their app to work on-premises and on the cloud and they have the security platform automatically configured as they deploy their apps with MapR.”
One of the underlying concepts behind the platform is that it serves as a multi-tier data system “that lets you define policies about what data goes where, and then allows you to leave the system alone and let it do its thing,” Dawar said. “The platform handles security and moves the data seamlessly without apps crashing while data management remains transparent and accessible, for the entire life cycle of the data.”
Other new security features MapR’s platform offers include file-based services for corporate security compliance with NFSv4 and volume-based data encryption as an additional layer to prevent unauthorized access to sensitive data and to help protect data in the event of packet-sniffing and storage device attacks.
MapR also says it has improved its support for Amazon Cloud Services (AWS) by adding S3 API support. Indeed, much of MapR’s improved analytics capabilities is designed for the S3 Interface for direct analytics on operational data, as well as for on-premises and multi-cloud environments.
Different cloud providers, such as Amazon, IBM, Google and Oracle, provide different database management capabilities similar to those on offer by MapR. However, the different alternatives, of course, are not necessarily designed to accommodate cloud deployments from other vendors.
“We accommodate the deployments from different providers with one solution,” says Bill Peterson, MapR’s vice president of industry solutions. “Each provider also might offer you 10 different products to do what we offer with a single platform.”
Indeed, as mentioned above, the main feature of MapR’s platform is that it serves as a common database and storage platform that supports various cloud and on-premises storage deployments—and is the only standalone offering of its kind on the market, notes Holger Mueller, analyst for Constellation Research. “MapR is in a unique position, as they choose to build their own storage system right at the inception of their database. This allows them to do interesting things around security and next-generation applications such as artificial intelligence, giving them some unique opportunities.”
