MapR Technologies has added new features in the MapR Converged Data Platform, including the ability to run stateful, containerized applications, such as those in Docker. In addition, the updated platform leverages Apache Myriad to enable sharing of data center resources between YARN and non-YARN jobs, and also provides additional security, data governance, and performance enhancements.
Support for Stateful, Containerized Apps
To go beyond transient, stateless, containerized applications, distributed persistent storage is required to complement Docker. The MapR Converged Data Platform acts as a data services layer for Docker containers, providing distributed, resilient storage for these containers, as well as the database and messaging/streaming capabilities that many containerized operational applications require.
“There is a big issue today, if you look at using Docker in the enterprise, in that many organizations are limiting that to only certain applications - ones that don’t require state and don’t requirement storage access,” said Jack Norris, SVP data and applications, MapR.
According to MapR, the new POSIX Client presents a distributed read-write file system to Docker containers for resilient Docker deployments on commodity hardware, allowing organizations to deploy data-oriented applications in Docker with the assurance that critical data will be persisted across application or server failures, or container movement across servers with no manual intervention.
In addition, to support for multi-tenant containerized environments, the new release includes Apache Myriad to enable infrastructure consolidation by sharing all data center resources between YARN and non-YARN jobs.
“MapR is taking our Converged Data Platform and providing a consistent data layer that can support the movement of containers throughout the enterprise and throughout on-premise and cloud environments,” said Norris. “It is a really transformational aspect of the Converged Data Platform. Now, organizations have much more flexibility as to which applications to put within containers – without having to worry about the physical and time-consuming administrative tasks to support them as they move.”
The new capabilities will be beneficial for a broad array of analytical and operational applications, said Norris. “Everything from customer-facing applications that are supporting an online environment, to production applications that are part of the supply chain process, to risk applications that are used to combat fraud and security compromises – any of those applications can now be deployed in containers.”
Beyond the new Docker support the 5.1 release of the MapR Converged Data Platform release also adds new big data capabilities across MapR Streams, MapR-DB, and MapR-FS. In particular there are improvements for security, governance, and performance, said Norris.
Security and Governance
In the new release of the MapR Converged Data Platform, file and stream access control expressions (ACEs) simplify the granting of permissions to users and groups across data files and directories using Boolean expressions, making security administration more scalable and manageable. Whole volume ACEs adds another level of protection for data files in MapR Volumes (a feature of MapR clusters), and provides greater multi-tenancy controls to guarantee that data is only available to specific groups. This is especially useful in hosted customer-facing SaaS applications to ensure that no client can access another customer’s information. In addition, selective auditing provides flexibility to track only the required activities to audit and/or analyze, giving flexibility in auditing while optimizing system performance.
“If you look at what we are providing with security, a big part of it is providing access control to the data,” said Norris. “Traditionally, that has been enabled through access control lists, but with big data there is a lot of complexity. What we have done with Access Control Expressions is introduce a powerful, but simple approach. It can be combined with groups - but you don’t have to do special one-off groups to accommodate some of the issues.”
“In terms of performance, the goal with the new release is to future-proof our platform so that organizations can get the maximum throughput from SSD deployments,” said Norris. According to MapR, an independent research firm, ESG, performed benchmark testing on MapR Streams and confirmed over 18,000,000 messages/second performance. In addition, MapR-DB, now with native JSON support, is performance optimized for SSDs with parallelized I/O access to NoSQL data for real-time operations, and the new POSIX Client provides up to 6GB/second client performance for file-based applications.
The MapR Converged Data Platform is available and free within the MapR Converged Community Edition. Cloud-based deployments for the new MapR Platform will be available in major public clouds, including Amazon Web Services (AWS) Marketplace, Azure Marketplace, and CenturyLink Marketplace.
More information is available from MapR.