CoreOS Inc. has moved closer to its goal of providing “Google infrastructure for everyone” with the release of a new image registry that’s designed to accelerate container builds on its software platform.

Container registries provide a way to securely store private container images, and the new Quay container registry brings a second Kubernetes-based component to its Tectonic container orchestration service, which is used to combine CoreOS’s stack with Kubernetes.

The new registry was unveiled by CoreOS at the ContainerCon Europe event in Berlin on Wednesday. The company provided further details in a blog post, saying that it’s able to leverage Kubernetes’ automatic scaling capabilities to speed up container builds by up to 80 percent on Tectonic clusters.

One of CoreOS’s central aims is to deliver web-scale infrastructure to its customers, and so all builds on its Quay image registry will be executed on Tectonic, which runs on bare metal servers provided by the infrastructure startup Packet.

“As container based stacks become more commonplace and complex, services like Quay play an increasingly critical role in everything from developer velocity to production stability and security,” Packet Chief Executive Zachary Smith said in a statement.

CoreOS said the new registry is able to transform source code, and the instructions that are used to automate the building of Docker images, called Dockerfiles, into complete container images. As a result, those who use the Quay.io registry will benefit from startup times of container builds of around 15 seconds. That compares with the three minutes it takes to boot up Amazon Web Services’ Elastic Compute Cloud instances, the company said.

Before, CoreOS relied on Amazon’s EC2 API to perform scheduling for each build, but now the Quay build managers uses Kubernetes’ API to schedule each job. In addition, using Kubernetes means computing resources can allocated more efficiently to individual builds.

CoreOS is planning to add support for enterprise Quay users soon, which will provide them with the ability to scale build clusters automatically, based on demand. That will do away with the requirement for manual scaling, the company said.

Image credit: Skeeze via pixabay.com