The pace at which technology moves can often be blindingly fast. A lifetime ago in tech years, version control wasn’t a particularly pressing consideration for organizations and their information technology departments.

Today, in addition to it being a core concentration area, with tools such as GitHub, advancements are already being made to facilitate continuous deployment for cloud-native applications. This paradigm is known as GitOps, and it is taking the enterprise computing world by storm.

At its core, the idea involves having a Git repository containing declarative descriptions at all times: firstly, of what is desired in the production environment and, secondly, of an automated process to bring the production environment to parity with the repository’s described state. It is, in essence, an automated way to manage applications in production.

“In a sense, it is very much aligned with DevOps, because we want to bring teams together and help them to deploy their applications, clusters and environments,” said Steve George (pictured, left), chief operations officer of Weaveworks Inc. “And really with GitOps, we have a specific set of tools that we can use. What’s nice about Git is … lots of developers use it, the vast majority. And so what we’re trying to do is bring those operational processes into the way that developers work.”

Rather than supplanting it entirely, GitOps is simply an instantiation of DevOps, according to George. Basically, it is using Git to record what is required in a runtime environment and then instructing Kubernetes, referencing the configuration in Git, on what should be deployed.

George and Steve Waterworth (right), technical marketing manager at Weaveworks, spoke with David Nicholson, host of theCUBE, SiliconANGLE Media’s livestreaming studio, during the AWS Startup Showcase: Open Cloud Innovations event. They discussed Weaveworks’ Kubernetes underpinnings, its connection with Amazon Web Services Inc. and how these have combined to bring forth the GitOps computing wave. (* Disclosure below.)

The AWS/Weaveworks connection

Amazon Elastic Kubernetes Services, a managed service allowing organizations to run and scale their K8s applications on AWS, has been developed in close partnership with Weaveworks. Alongside the two companies’ collaborative efforts around Kubernetes, AWS is also an investor in Weaveworks, according to Waterworth.

Earlier, during EKS’ early days, Weaveworks helped conceptualize and develop its command line interface and ensure that its primary user base, the developer community, found it intuitive and easy to work with. Leveraging this connection, both companies devised ways to integrate the GitOps approach into EKS. Because it’s all open source, community members can get in on the action and provide their much-needed feedback and contributions.

With a developer-centric operating model for Kubernetes, GitOps consolidates all of the commonly used developer services and systems into a familiar base, reducing the burden on them to switch between different tools.

“For operations staff, one of the things that it does is it centralizes where all of this configuration is kept. And then you can use things like templating to make sure that you automate and go quickly, but you also do that in a way that is reliable, secure and stable,” George explained.

GitOps is a core component of Weaveworks’ Kubernetes management value proposition. Thus, the company’s technology has several in-built features that help organizations key into the GitOps paradigm.

“Weaveworks has an engine, a set of software, that enables this to happen,” Waterworth explained. “Think of it as a constant reconciliation engine. So, you’ve got your declared state; your desired state is declared in Git. So, this is where all your YAML for all your Kubernetes hangs out.”

The Weaveworks GitOps agent runs inside of Kubernetes and constantly compares the desired Git state with the actual state, within the container. Whenever a developer wants to make a change, they push that change to Git, and the reconciliation loop does the required comparisons.

Now, here’s where the real magic happens. Engineers and developers need not learn any new changes. The whole process functions the same as the usual action of pushing code into Git.

Weaveworks and trusted application delivery

With cloud spending constantly rise, organizations are increasingly wary of adding new “ops” to their IT processes, because they often imply diverting resources towards infrastructure. However, GitOps adds an increased level of efficiency to infrastructure deployment and usage.

“In the GitOps pipeline, one of the things that we can do is to put in various checkpoints to check that the policy is being followed correctly,” George said. “So, are we deploying the right number of applications, the right configuration of an application? Does that application follow certain standards that the enterprise has set down? And that’s what we talk about when we talk about trusted policy and trusted delivery.”

Trusted policy and delivery will allow teams to work at high speeds with their new deployments, but also safeguard the work with “automated guard rails” to maintain the application platform’s reliability.

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of the AWS Startup Showcase: Open Cloud Innovations event. (* Disclosure: Weaveworks sponsored this segment of theCUBE. Neither Weaveworks nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE