Cognitive

Building a cloud for AI with Kubernetes, DevOps and scale

Share this post:

Since the launch of the IBM Cloud Kubernetes Service, our goal has been to provide developers with the broadest and easiest range of ways to scale, run and build with containers.

Containers are natural building blocks for the next era of cloud workloads, such as AI and machine learning. We know the benefits of fueling AI with containers firsthand: our own Watson services run on Kubernetes.

The secret to running well-tuned AI lies in containers. When designed correctly, containers offer instant connections into AI tools and address concerns around security, scale and infrastructure that can distract developers from focusing on improving their code.

This is why we are expanding their ease and flexibility even further. Today, we are launching the availability of multi-zone clusters for our IBM Cloud Kubernetes Service. Now, our users can deploy containerized applications across multiple global IBM Cloud zones within a region as a single operation.

With multizone support, the IBM Cloud Kubernetes Service manages each piece of a containerized app simultaneously across different zones, such as rolling out updates, connecting it into the high-value services the IBM Cloud offers, and finding vulnerabilities.

This multi-zone capability is available across multiple regions all over the world, including the IBM Cloud’s 18 new IBM Cloud availability zones, announced this week at CeBIT in Germany. Developers will be able to easily deploy containerized apps anywhere they need them. These capabilities simplify how developers deploy and manage containerized applications, and add consistency to their development experience.

Increasing the availability, scale and ease of containers ties into other advancements we’ve made to prime the IBM Cloud Kubernetes Service as the ideal choice for AI. Earlier this year, we launched the option to run our Kubernetes service on bare metal and with GPUs, enabling developers to access the high performance power of GPUs that AI and machine learning workloads often require.

We are also continuing to forge partnerships to accelerate the agility and intelligence with which developers can build container-based apps. DevOps is in our DNA: we update our Kubernetes service up to 100 times each day, and we want our users to be able to operate with this same speed.

Rapidly growing startups Sysdig, a cloud-native intelligence provider, and LogDNA, a container logging and log analyzing platform, will join the IBM Cloud partner ecosystem to help us achieve this goal. Together, we’ll be working to bring developers new levels of intelligence into what’s happening in their containers. Equipping teams with this insight helps to free them from many of the time-consuming tasks necessary to manage infrastructure, and positions them to do more of what they love: iterate, build new features, and code.

Expanding our IBM Cloud Kubernetes Service takes care of underlying infrastructure concerns that must be in place to build and scale AI, blockchain and Internet of Things (IoT). We’ve already made great strides forward to take steps to lessen the pressing burdens on developers, such as our work with Google and Lyft to build Istio, which helps orchestrate, secure and manage traffic and data flowing through containers and microservices.

As container use grows and they continue to make their case as the defacto building blocks of cloud and AI, we will continue progressing their security and stability for enterprises.

Learn more about how we’re helping bring containers to industries and businesses across the globe.

More Cognitive stories

Continuous delivery and the DevOps approach to modernization

Businesses are working to increase agility, deliver innovative and engaging experiences to clients, and stay ahead of competition. Increasingly, companies are modernizing business applications to make these business goals a reality. Modernizing applications is generally composed of three transformations: cloud-native architecture, continuous delivery and infrastructure automation. These typically occur concurrently, but do have distinct characteristics. […]

Continue reading

Clarifying cloud personalization: 4 common misconceptions

You wouldn’t purchase a car without some forethought, and you shouldn’t rush into any cloud environment without careful consideration. Your organization undoubtedly has widely varying IT needs, so relying on a one-size-fits-all cloud computing solution would be similar to driving off the lot with a sports car when you really need an SUV. Cloud personalization […]

Continue reading

How DevOps tools make applications go faster and farther

The term DevOps has been in existence for at least a decade now. In tech years, a decade is a long time. It’s hard to measure how pervasive the DevOps model has become in that decade, but it’s safe to say that most, if not all, development and operations teams know about it, and many […]

Continue reading