We've said it before, and we'll say it again: the DevOps mode of software development is fast becoming one of the new big forces in the channel. Here's a look at some of the key projects and products in the open source DevOps space, and an explanation of how each one will change the way organizations create and VARs integrate software.
Ansible is a configuration management platform that promises to deliver "infrastructure as code." What that means is that admins can create configuration files that define how servers, networks, storage and applications should behave. Ansible then implements the policies across the infrastructure, even if it is comprised of diverse parts. The result is software that can be managed with minimal human interaction and little reliance on configuration tools that work with only a specific type of hardware or software.
Ansible and comparable tools, like Puppet, are reshaping administrative services and the relationship between software and hardware. Vendors who offer service contracts need to adapt for an age in which users expect to be able to manage all of their infrastructure from a centralized location, without paying extra for proprietary services or configuration tools.
CoreOS is building an increasingly sophisticated ecosystem of software tools for creating, deploying and managing apps through containers. CoreOS's containers are different from Docker containers in a technical sense, although the really important distinction between Docker and CoreOS is that the latter is shooting to offer a comprehensive and integrated ecosystem (kind of like Canonical has done for Linux by building a broad ecosystem of home-grown tools around Ubuntu). For now, Docker is focusing mainly on containers, not the tools for deploying them.
For the channel, the big reason to watch CoreOS is to see how much adoption its container platform enjoys. If it proves popular, it will signal interest in one-stop container solutions, as opposed to Docker infrastructures, which generally include parts from multiple vendors.
Docker is probably the best-known name in the DevOps world at the moment. The company didn't invent containers, but it took them mainstream by showing developers and admins how easy it can be to create code for containers, place binaries inside container environments and then deliver containers quickly to users, no matter what kind of infrastructure they're using.
More than any other project to date, Docker has made portable applications an expectation for many organizations. By extension, it has eased software integrations by making apps more environment-agnostic than ever. Using containers, partners can readily collaborate without having to tweak their products in order to make them compatible.
Jenkins has emerged as one of the leading Continuous Delivery platforms. Continuous Delivery means that the processes of writing, testing, building and delivering code do not occur in distinct stages at fixed intervals. Instead, these processes are continually in sync with one another, and they occur on an ongoing basis.
The idea behind Continuous Integration is to make software production more efficient by eliminating the barriers between different teams. Jenkins facilitates this goal by allowing users to write "pipe lines" using software scripts, which automate the Continuous Delivery process.
For VARs, Continuous Integration means that users now expect software to be released earlier and more often than ever, without compromising on quality. The era when you could get away with updating your products once ever few years is over.
Kubernetes, which originated at Google, is designed to solve the considerable challenge of managing a large number of container-based apps running across a cluster of servers. That's important because the management tools from the container vendors themselves, like Docker and CoreOS, are not designed to automate container management or make it work well at scale.
Kubernetes and Docker Compose, Docker's comparable tool, underline the importance of scale in today's software world. Users no longer expect to be limited by the size of staffs or the available of work hours for deploying large infrastructures. They want to be able to run ten thousands containers as easily as they can run a dozen.
OpenShift, a Red Hat platform, extends the functionality of Kubernetes by integrating container orchestration with hardware and software in order to deliver a Platform-as-a-Service, or PaaS. It's one of the first major examples of a fully integrated product for running container-based apps.
While OpenShift is built primarily with Red Hat tools, they're all open source, creating opportunities for VARs who think they can do even more than Red Hat already is with the product. OpenShift also sets an example of what a PaaS in the DevOps space can be, and will likely prompt other vendors to seek their own integrated solutions for container deployment.
Terraform promises to expand the value of the infrastructure-as-code principle by enabling the integration of any type of infrastructure, even if it includes parts from different vendors whose products were not designed to work together.
Terraform is an answer to the challenge of assuring backwards-compatibility as new DevOps tools emerge. That's significant since, while it would be nice if all infrastructure suddenly became as open as DevOps tools themselves, that's unlikely to happen. Hooking into legacy systems and bringing them up to speed with DevOps workflows will be key as DevOps adoption continues.
January 31: 5 Ways to Transform Your MSP Service Delivery Using IT Automation
Watch Now: How to Help Your Customers Protect Themselves from Ransomware Attacks
For VARtweet sponsor info contact Amy Katz (email@example.com)
Sponsored Introduction Continue on to (or wait seconds) ×