Dataware for data-driven transformation

MapR Clarity Program: The Clear Path to Containers - Video

8 min read

Editor's Note: In this video interview, Bill Peterson and Suzy Visvanathan talk about the recently announced MapR Clarity Program. Watch this interview to learn how MapR is a leader and early adopter of both Docker and Kubernetes, and has made containerized stateful applications a reality.

  • Build for performance on the best platform for AI and ML today.
  • Get the portability and scale of containers today; no need to wait.

Learn more about the MapR Clarity program or how MapR compares to Cloudera.

If you’re interested in learning about what the MapR Clarity Program can do for you, watch the full on-demand webinar here.

Video Transcript

Hi, I'm Bill Peterson. Vice President of Industry Solutions. I'm here today with.

Suzy Visvanathan, Director of Product Management at MapR

Great and today we're gonna talk about containers and containers in regards to our recently announced Clarity Program. Suzy, what did we announce with Clarity in regards to containers for Cloudera and Hortonworks customers?

With MapR we have been continuously investing in containers and Kubernetes in general as well so we have had a long history with producing features and allowing customers to run containerized apps on MapR. With Clarity Program, what we intend to do is bring the benefits of that in a very easy manner for customers to move their existing environment to a containerized environment. We are also focusing on just not about just the already done features but also focusing on what we'll bring to them in the future as well. We have a heavy investment going on in the containers and Kubernetes area. With this Clarity Program and the StepUp Program within that, customers will be able to understand more about the benefits of containers on MapR and also how our long-term vigilant strategy on that will benefit them.

Makes sense. What's better about our approach to containers than Cloudera or Hortonworks?

Our foundational platform, if you will, is residing on the fact that we are built for scale. We are built for performance. Based off of that we are building the best platform for AI and ML today. We also have a great integration with NVIDIA GPU workloads for instance. What we have differentiated ourselves when it comes to containers especially is we want to take this scale. We want to take this distributed environment where you can run different analytic tools, different AI and ML tools and how do you containerize them? We already have all the recipe for it. We already have the ingredients for it. It's a matter of combining that with the benefits of containers.

What does the benefit of containers bring you? It brings you portability. It brings you an easy way to spin instances and delete them if you want. It brings you the ability to scale computing dependent of storage. All those governors of containers the docker has given us or the open source has given us combined with our platform, benefits security, scale, performance. All of this combined together will give a whole end to end solution for customers.

Got it. You mentioned Kubernetes in there. What are we doing with Kubernetes or some of the others to manage containers?

Earlier this year we gave a way for customers to run stateful containerized apps on MapR. They can continue to use MapR as a separate cluster from the Kubernetes cluster. Thereby, we will give them persistent storage. We not only are doing that today. Even prior to that, we gave them an easy way to run containerized app using our POSIX client. A POSIX client is what ensures that you get the max throughput and bandwidth on a MapR environment. We have already taken steps towards integrating completely with Kubernetes. Now we are thinking about not just Kubernetes but everything that comprises of the CNC app. This is what I was referring to as it's not just what we have done, but also where our vision and strategy lies.

As you follow us, you will see how we are investing in this and if something that customers are interested in, we always have pretty good beta programs where they can get the enhancement on what they can do with that product and what's coming in the future. That is something that we are definitely throwing out for customers. If they are interested in this, I'm open to actually getting more prospects and customers to come and take a look at what we have done.

On the subject of customers for our viewers, let's put a little context around it. Give us just a brief example of a MapR customer today and what they're doing with containers.

Customers fall into two categories for us when it comes to containers. One is where they are trying to spin up new apps. I already mentioned AI ML. That gives them an easy way to actually spin up new environments with containers because old legacy applications have in-built requirements and in-built dependency on an existing infrastructure. Customers who want to have say, deep learning. I'm talking about deep learning because that is a form of machine learning. What I mean by that is, if an app is just doing facial recognition, that is an example of machine learning. If an app keeps going to the first round layer of information then digs deeper into the second layer of information, that is an example of deep learning.

Customers, one I would cite a common customer profile of this who does this, they are into a big set of customers fall under autonomous car solutions. They are classic examples of deep learning. They take car data and they drill deeper, and deeper, and deeper and learn from it. They predominantly will benefit from our Kubernetes environment from a containers environment. That is one segment of customers. The second segment I'm actually seeing are edge customers. Edge used to be associated with IoT devices, but it's become a lot more than that. Edge nodes, edge clusters have become self-sufficient where they want to be able to get the value out of the data at the edge nose, at the edge clusters without just being a dummy source of information and sending it to the data centers. In those cases, containers Kubernetes is an excellent deployment method because it consumes less resources. It can be self-sufficient. You can scale up a cluster to get the data from and then immediately dismantle it and you don't have the headache of having to maintain an environment at the edge.

Great. Great stuff. Thank you very much. Join us next time and we'll talk about MapR and real time analytics.

Learn more about the MapR Clarity program or how MapR compares to Cloudera.


This blog post was published December 17, 2018.
Categories

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