New container image deployment method for Red Hat Enterprise Linux is aimed at helping enterprises streamline operations and management.
At the Red Hat Summit this week, the company unveiled a new container image deployment method for Red Hat Enterprise Linux. The new option is designed to streamline operations, enhance consistency across hybrid cloud environments, and accelerate the adoption of cutting-edge technologies like AI and machine learning.
Typically, containers trim down operating systems as much as possible because they run within a host OS, says Bradley Shimmin, chief analyst for AI platforms, analytics, and data management at Omdia. Or Linux is run within virtual machines, also running on top of an underlying operating system. But this creates management complexity.
“Red Hat is using the Open Container image standard to create bootable images, which look and work just like it were the actual OS running on bare metal,” says Shimmin.
Now enterprises can use all the tools that they already have in place for managing containers. “Containerization is the current paradigm that everyone accepts and values, frankly, for deploying software for any kind and any sort,” Shimmin says.
(Related news from Red Hat Summit: Red Hat extends Lightspeed genAI tool to OpenShift and introduces ‘policy as code’ for Ansible)
The idea of putting Linux in a container isn’t new. There are community projects that deliver bootable containers, including Bluefin and Fedora. Even Red Hat’s Linux has been available as an image previously, such as the Red Hat Universal Base Image. In addition, Red Hat also has a container-based operating system for Red Hat OpenShift in Red Hat Enterprise Linux CoreOS.
“But image mode for RHEL is one of, if not the, first enterprise Linux platform to offer it,” says Ben Breard, senior principal product manager at the Red Hat Enterprise Linux business unit.
“The entire operating system will be delivered as a bootable container,” Breard says. “Universal Base Image still needed to be run on a host operating system, for example – it was a container on a host.”
The advantage to doing this is that it can help enterprises streamline operations and management, maintain a consistent and reliable infrastructure whether on bare metal, on virtual machines, or in public clouds.
Companies rarely use an out-of-the-box operating system, Breard says. Instead, they build a standard operating environment by layering in hundreds, or thousands, of additional packages to meet their specific needs. “Problems arise when patches, updates and upgrades have to be pushed out,” he says. “Making changes to the underlying image can be incredibly tedious, time consuming and complex.”
Old-school monolithic applications had the same problem, he says. “But along came containers, which enabled discrete pieces of the app to be packaged and updated individually. So, applying this type of methodology to gold images would be a huge timesaver and innovation driver for enterprise IT. This is the initial problem that image mode solves.”
Being able to make these changes quickly is even more crucial for artificial intelligence, he says. “Patches and upgrades need to be pushed – and work – immediately,” he says. “If you can’t go fast, you can’t reap the benefits of AI workloads.”
One new capability that image mode makes possible, which will accelerate things even further, is that operations teams will now be able to use the same container tooling and workflows as developers, he says.
“This means that changes can be pushed to standard operating environments much faster, and it enables technology organizations to standardize on tooling across both the operations and developer teams,” he says. “And it makes it even easier to push updates out to customers with vast IT estates – a single patch or driver update can be pushed from a single console via a container, and everything is installed and updated via container magic.”
Users will also be able to view and update image mode deployments directly from Red Hat Insights. That will give companies a smarter approach to risk management, says Gunnar Hellekson, vice president and general manager of the Red Hat Enterprise Linux business unit.
“This is the proactive advice feature of the RHEL subscription,” Hellekson says. “We’re able to tell people ahead of time what their CVE exposure might be when they put together a certain kind of image. This is part of an overall move that we have to improve the amount of intelligent coaching and help that we can offer customers during the build and construction of RHEL, as opposed to after RHEL is deployed.”
Additional security benefits come from the fact that security teams will now be able to apply container security tools such as scanning, validation, cryptography and attestation to the base elements of the operating system.
