Beyond Kubernetes Container Orchestration

on Mar 23, 2017

If you’re going to successfully deploy containers in production, you need more than just container orchestration

Kubernetes is a valuable tool

Kubernetes is an open-source container orchestrator for deploying and managing containerized applications. Building on 15 years of experience running production workloads at Google, it provides the advantages inherent to containers, while enabling DevOps teams to build container-ready environments which are customized to their needs.

The Kubernetes architecture is comprised of loosely coupled components combined with a rich set of APIs, making Kubernetes well-suited for running highly distributed application architectures, including  microservices, monolithic web applications and batch applications.  In production, these applications typically span multiple containers across multiple server hosts, which are networked together to form a cluster.

Kubernetes provides the orchestration and management capabilities required to deploy containers for distributed application workloads. It enables users to build multi-container application services and schedule the containers across a cluster, as well as manage the health of the containers.  Because these operational tasks are automated, DevOps team can now do many of the same things that other application platforms enable them to do, but using containers.

But configuring and deploying Kubernetes can be hard

It’s commonly believed that Kubernetes is the key to successfully operationalizing containers at scale.  This may be true if you are running a single Kubernetes cluster in the cloud or have reasonably homogenous infrastructure. However, many organizations have a diverse application portfolio and user requirements, and therefore have more expansive and diverse needs. In these situations, setting up and configuring Kubernetes, as well as automating infrastructure deployment, gives rise to several challenges:

  1. Creating a Kubernetes environment that is customized to the DevOps teams’ needs
  2. Automating the deployment of multiple Kubernetes clusters
  3. Managing the health of Kubernetes clusters (e.g. detecting and recovering from etcd node problems)
  4. Automating the upgrade of Kubernetes clusters
  5. Deploying multiple clusters on premises and/or across disparate cloud providers
  6. Ensuring enterprise readiness, including access to  24×7 support
  7. Customizing then repeatedly deploying multiple combinations of infrastructure and other services (e.g. storage, networking, DNS, load balancer)
  8. Deploying and managing upgrades for Kubernetes add-ons such as Dashboard, Helm and Heapster

Rancher is designed to make Kubernetes easy

Containers make software development easier by making code portable across development, test, and production environments. Once in production, many organizations look to Kubernetes to manage and scale their containerized applications and services. But setting up, customizing and running Kubernetes, as well as combining the orchestrator with a constantly changing set of technologies, can be challenging with a steep learning curve.

The Rancher container management platform makes it easy for you to manage all aspects of running containers.  You no longer need to develop the technical skills required to integrate and maintain a complex set of open source technologies. Rancher is not a Docker orchestration tool—it is the most complete container management platform.

Rancher includes everything you need to make Kubernetes work in production on any infrastructure, including:

  • A certified and supported Kubernetes distribution with simplified configuration options
  • Infrastructure services including load balancers, cross-host networking, storage drivers, and security credentials management
  • Automated deployment and upgrade of Kubernetes clusters
  • Multi-cluster and multi-cloud suport
  • Enterprise-class features such as role-based access control and 24×7 support

 

 

We included a fully supported Kubernetes distro

The certified and supported Kubernetes distribution included with Rancher makes it easy for you to take advantage of proven, stable Kubernetes features.  Kubernetes can be launched via the easy to use Rancher interface in a matter of minutes. To ensure a consistent experience across all public and private cloud environments, you can then leverage Rancher to manage underlying containers, execute commands, and fetch logs.  You can also use it to stay up-to-date with the latest stable Kubernetes release as well as adopt upstream bug fixes in a timely manner.  You should never again be stuck with old, outdated and proprietary technologies.

The Kubernetes Dashboard can be automatically started via Rancher, and made available for each Kubernetes environment. Helm is automatically made available for each Kubernetes environment as well, and a convenient Helm client is included in the out-of-the-box kubectl shell console.

We make Kubernetes enterprise- and production-ready

Rancher makes it easy to adopt open source Kubernetes while complying with corporate security and availability standards. It provides enterprise readiness via a secure, multi-tenant environment, isolating resources within clusters and ensuring separation of controls. A private registry can be configure that is used by Kubernetes and tightly coupled to the underlying cluster (e.g. Google Cloud Platform registry can be used only in a GCP cluster, etc.).

Features such as role-based access control, integration with LDAP and active directories, detailed audit logs, high-availability, metering (via Heapster), and encrypted networking are available out of the box. Enterprise-grade 24x7x365 support provides you with the confidence to deploy Kubernetes and Rancher in production at any scale.

Multi-cluster, multi-cloud deployments? No problem 

Kubernetes eBook

Quickly get started with Rancher and Kubernetes by following the step-by-step instructions in the latest release of the Kubernetes eBook.

Rancher makes it possible to run multi-node, multi-cloud clusters, and even deploy stateful applications.  With Rancher, Kubernetes clusters can span multiple resource pools and clouds. All hosts that are added using Docker machine drivers or manual agent registration will automatically be added to the Kubernetes cluster. The simple to use Rancher user interface provides complete visibility into all hosts, the containers running in those hosts, and their overall status.

But you need more than just container orchestration…

Kubernetes is maturing into a stable platform.  It has strong adoption and ecosystem growth.  However, it’s important not to lose sight that the end goal for container adoption is to make it easier and more efficient for developers to create applications and for operations to manage them.  Application deployment and management requires more than just orchestration.  For example, services such as load balancers and DNS are required to run the applications.

Customizable infrastructure services

The Rancher container management platform makes it easy to define and save different combinations of networking, storage and load balancer drivers as environments. This enables users to repeatedly deploy consistent implementations across any infrastructure, whether it is public cloud, private cloud, a virtualized cluster, or bare-metal servers.

The services integrated with Rancher include:

  • Ingress controller with multiple load balancer implementations (HAproxy, traefik, etc.)
  • Cross-host networking drivers for IPSEC and VXLAN
  • Storage drivers
  • Certificate and security credentials management
  • Private registry credential management
  • DNS service, which is a drop-in replacement for SkyDNS
  • Highly customizable load balancer

If you choose to deploy an ingress controller on native Kubernetes, each provider will have its own code base and set of configuration values.  However, Rancher load balancer has a high level of customization to meet user needs. The Rancher ingress controller provides the flexibility to select your load balancer of choice—including HAproxy, Traefik, and nginx—while the configuration interface remains the same. Rancher also provides the ability to scale the load balancer, customize load balancer source ports, and schedule the load balancer on a specific set of hosts.

A complete container management platform

You’ve probably figured this out for yourself by now but, to be clear, Rancher is NOT a container orchestrator.  It is a complete container management platform that includes everything you need to manage containers in production. You can quickly deploy and run multiple clusters across multiple clouds with a click of the button using Rancher or select from one of the integrated and supported container orchestrator distributions, including Kubernetes as well as Mesos,Docker Swarm and Windows. Pluggable infrastructure services provide the basis for portability across infrastructure providers

Whether running containers on a single on-premises cluster or multiple clusters running on Amazon AWS and other service providers, Rancher is quickly becoming the container management platform of choice for thousands of Kubernetes users.

 

Get started with containers, Kubernetes, and Rancher today!

Ready to give Rancher and Kubernetes a try? Check out the Rancher Sandbox at try.rancher.com. Just sign in with GitHub, and add your private or public hosts to get started.

For step-by-step instructions on how to get started with Kubernetes using the Rancher container management platform, please refer to the Kubernetes eBook, which is available here.  Or, if you are heading to KubeCon 2017 in Berlin, stop by booth S17 and we can give you an in-person demonstration.

 

 

Louise is the Vice President of Marketing at Rancher Labs where she is focused on defining and executing impactful go-to-market strategy and marketing programs by analyzing customer needs and market trends. Prior to joining Rancher, Louise was Marketing Director for IBM’s Software Defined Infrastructure portfolio of big data, cloud native and high performance computing management solutions. Before the company was acquired by IBM in 2012, Louise was Director of Marketing at Platform Computing. She has 15+ years of marketing and product management experience, including roles at SGI and Sun Microsystems. Louise holds an MBA from Santa Clara University’s Leavey School of Business and a Bachelor’s degree from University of California, Davis. You can follow Louise in Twitter @lwestoby.

Free Guide: Choose Kubernetes, Mesos, or Swarm

Get a detailed comparison

Download an in-depth guide comparing Kubernetes, Mesos, and Docker Swarm, along with key criteria to consider for your use case.

Free eBook: CI/CD Pipelines with Docker and Rancher