Ansible simplicity and the shift to AI automation

The history of Red Hat and Ansible

I see the acquisition of Ansible as a signal that Red Hat wants to lead in hybrid cloud management. Ansible started in 2013 in Santa Barbara and raised $6 million from Menlo Ventures and partner Doug Carlisle. The company was founded by Michael DeHaan, who worked in the research and development arm of Red Hat. Red Hat announced it would buy Ansible for approximately $150 million, though rumors suggested a price of over $100 million. This acquisition follows IBM’s $34 billion purchase of Red Hat in 2018. Ansible built a reputation through partnerships with Cisco, HP, CSC, and RackSpace to support OpenStack clouds. Red Hat expects the acquisition to increase its non-GAAP operating expenses by $2 million in the third quarter and $4 million in the fourth quarter of fiscal 2016. It also helps Red Hat compete with VMware, Microsoft, and IBM. I recommend Ansible if your team needs to manage applications across private and public clouds or wants to simplify container adoption. You can use Ansible to manage Linux, Windows, macOS, Docker, and Kubernetes.

Architecture and configuration management

I find that the architecture of these tools determines the workload for your IT team. Ansible uses SSH to connect to nodes, which makes it the easiest option to configure. You only need to set up a primary instance and create an SSH connection, as you already know how to handle remote access. Chef requires you to configure a main server and a backup server, then install an agent on every managed server. Puppet also requires a main server and agents on node servers, plus you must configure certificate signing between the server and the agents. Because Ansible relies on SSH for connections, its architecture can be slow when managing thousands of nodes, which makes agent-based tools like Puppet a better choice for massive, high-scale deployments.

Ansible is written in Python and works by pushing Ansible modules to managed nodes. It uses idempotent operations, which means it only makes changes when they are needed. You can use it for rolling updates, blue-green deployments, and disaster recovery with conditional logic. Since it is an open source project, almost 1,200 outside developers contribute to it. Chef requires users to have programming skills to write Ruby recipes, whereas Puppet uses its own declarative language called Puppet DSL. Chef also needs a workstation to store configurations for testing before you push them to the main server. Puppet and Chef use a centralized approach where nodes pull configurations from a main server, but Ansible uses a decentralized approach where configurations are distributed across multiple servers. If you need to manage 10,000 or more nodes with continuous compliance, Puppet is the better choice because its agents check in every 30 minutes to fix configuration drift. Puppet provides tools like Puppet Remediate and Compliance Automation to help manage vulnerabilities and ensure nodes follow policies. Chef uses InSpec to define and automate security policies. Ansible uses Ansible Tower to manage security policies and can integrate with third-party vulnerability scanners. Chef and Puppet can manage cloud platforms like AWS, Google Cloud, or Microsoft Azure.

The future of automation and enterprise scale

I see the industry moving toward generative AI to help solve the skills gap in IT. Red Hat announced Ansible Lightspeed this week at its annual customer summit in Boston. This tool uses IBM’s Code Assistant and IBM’s foundational models, which are trained on the Ansible ecosystem, all playbooks, and subject matter expertise. When a user describes a workflow, Lightspeed creates it based on that description. The public version of Lightspeed helps train the data model, while the enterprise version arrives in October. Red Hat is taking advantage of its relationship with IBM to bring this generative AI product to market more quickly. Event-driven Ansible is also generally available to automate advanced developer or SRE tasks. I suggest using Ansible Tower if you need a professional GUI, role-based access control, and job scheduling. You can use the free AWX version if you want to test new features in an open-source environment, though AWX undergoes minimal testing and quality engineering. Do you need to manage a large, complex environment with strict compliance requirements? For small, consistent environments, the core Ansible tools are enough. The market for systems management software reached $8.3 billion in 2019, up from $2.3 billion in 2014.