MapR and Kubernetes: Do More with Big Data

Contributed by

7 min read

At the Strata Data Conference earlier this year, we announced that MapR customers will now be able to utilize Kubernetes containerization technology on the MapR Data Platform.

So, how does our data platform make the most of Kubernetes, why should our customers care, and what does the industry have to say about it? Let's look at Kubernetes, its opportunities, and challenges.

What Is Kubernetes?

Originally developed by Google, Kubernetes is a comprehensive open source containerization tool. At its core, Kubernetes provides the foundation for organizations to leverage their big data by deploying machine learning models and building progressive applications.

Designed to deploy, scale, and manage containerized applications, Kubernetes has a variety of key features, including:

  • Horizontal scaling: Effortlessly scale your application up and down.
  • Self-healing: With Kubernetes' ability to self-heal, you'll never have to worry about manually restarting, rescheduling, or killing containers.
  • Automatic bin packing: Efficiently and automatically places containers.

And, since Kubernetes was created with Google's massive amount of containers in mind, it's capable of handling even the most demanding loads.

However, Kubernetes, and containers in general, are not without their challenges. Jack Norris, our Senior Vice President of Data and Applications, explains those challenges in an interview with SDxCentral:

"Containers have made it harder in some cases to share data across an organization because if that container goes away, the data running inside of that container goes with it."

As a recent article from Datanami points out, another one of the challenges that comes with containerization is the persistency of stored data. In other words, containers don't always ensure that data is saved after it's been accessed and used.

MapR Data Fabric for Kubernetes

The MapR Data Platform uses an edge-to-cloud data fabric to store, manage, process, analyze, and apply data in real time.

Meanwhile, many enterprises are migrating their applications to containers. However, many container orchestration tools aren't able to provide data management and data protection.

That's where MapR's native Kubernetes integration comes in.

With MapR's Kubernetes integration, you can easily build a secure, reliable, and scalable containerized platform and stateful container applications.

In this video for Truth in IT, Norris discusses this new integration with contributing vlogger Mike Matchett:

MapR's Kubernetes integration allows you to enjoy:

  • Security and multi-tenancy: MapR tickets make it possible to achieve end-to-end security of containers accessed by various users and groups.
  • High availability: Thanks to automatic failover, data is always available and containerized applications can run 24 hours a day, seven days a week.
  • Data protection: Mirroring, replication, and point-in-time snapshots work together to protect your most important data.
  • Multiprotocol data persistence: Use POSIX, NFS, and S3 protocols to create, retain, and synchronize data volumes.
  • Extreme scalability: A convenient "hot add" option lets you scale the clusters as your number of containers increases.
  • High performance: Flexibly deploy on NVMe, SSDs, HDDs, or cloud to meet your performance SLAs.

"By containerizing applications and not tying them to specific servers, I can more easily automate jobs and get more out of my infrastructure because my utilization is higher," Norris tells SiliconANGLE. That is to say, "I can move an application across any of 30,000 servers and have the same access."

In an article, The New Stack sums up the purpose of our Kubernetes integration nicely:

"The idea is to allow developers to worry about what they do best: deploying apps on Kubernetes without having to fret over data access with persistent storage. That's because persistent storage now comes loaded and ready on MapR's platform."

MapR has always supported containers. However, with the help of Kubernetes, our customers are able to access persistent storage more easily than ever before.

Why Should MapR Customers Care About Kubernetes?

Yes, Kubernetes is an advanced containerization tool. And yes, MapR provides an edge-to-cloud data fabric. But why should our customers care about MapR's new native integration of Kubernetes?

As Norris tells Datanami, customers should be excited "because of the capabilities in the underlying platform." Even though "the update [that added support for Kubernetes] was light," it leverages the "near-decade of innovation in that platform, which are not small innovations," says Norris.

Additionally, MapR directly addresses the complexity of native Kubernetes support in a way that other vendors do not. Norris explains:

"What we've really done is address the drivers of that complexity with our underlying data fabric so that that access can be so straightforward and so simple and easy to integrate into that environment."

In fact, chances are that most MapR customers are already excited about our Kubernetes integration — one of the main reasons that we strove to provide it was the demand of our customers.

As our Director of Product Management Suzy Visvanathan puts it, "Kubernetes is popular among customers mainly because it scales a lot more than Docker, and [because] the deployment of it is integrated already with the main cloud vendors."

The Industry Reacts to MapR's Kubernetes Support

There's been overwhelming positive feedback to MapR's new Kubernetes support.

James Kobielus, lead analyst at SiliconANGLE Wikibon and Datanamicolumnist, says that:

"Stateful and data-driven applications can't elegantly live in the cloud without an elegant means for persisting state and making it available, securely and robustly, to containerized microservices. We're impressed with how MapR has addressed these challenges for stateful containers that run in big-data clusters that have deployed Kubernetes."

Additionally, Holger Mueller, Vice President and Principal Analyst at Constellation Research, told Digitizing Polaris that "there are only very few business functions that don't require persistence beyond their execution." The article continues: "He adds that it is good to see established enterprise vendors (like MapR) solve the problem which will, in turn, give more confidence to CxOs to deploy their next gen apps on Kubernetes and the respective databases."

With our new built-in Kubernetes support, we're excited to continue giving our customers the best data platform on the market.

This blog post was published August 07, 2018.

50,000+ of the smartest have already joined!

Stay ahead of the bleeding edge...get the best of Big Data in your inbox.

Get our latest posts in your inbox

Subscribe Now