In the beginning, there were sysadmins
Since the beginning of networked computing, deploying and managing servers reliably and efficiently has been a challenge. Historically, system administrators were walled off from the developers and users who interact with the systems they administer, and they managed servers by hand, installing software, changing configurations, and administering services on individual servers.
As data centers grew, and hosted applications became more complex, administrators realized they couldn’t scale their manual systems management as fast as the applications they were enabling. That’s why server provisioning and configuration management tools came to flourish.
Server virtualization brought large-scale infrastructure management to the fore, and the number of servers managed by one admin (or by a small team of admins), has grown by an order of magnitude. Instead of deploying, patching, and destroying every server by hand, admins now are expected to bring up new servers, either automatically or with minimal intervention. Large-scale IT deployments now may involve hundreds or thousands of servers; in many of the largest environments, server provisioning, configuration, and decommissioning are fully automated.
Modern infrastructure management
As the systems that run applications become an ever more complex and integral part of the software they run, application developers themselves have begun to integrate their work more fully with operations personnel. In many companies, development and operations work is integrated. Indeed, this integration is a requirement for modern test-driven application design.
As a software developer by trade, and a sysadmin by necessity, I have seen the power in uniting development and operations—more commonly referred to now as DevOps or Site Reliability Engineering. When developers begin to think of infrastructure as part of their application, stability and performance become normative. When sysad- mins (most of whom have intermediate to advanced knowledge of the applications and languages being used on servers they manage) work tightly with developers, development velocity is improved, and more time is spent doing ‘fun’ activities like performance tuning, experimentation, and getting things done, and less time putting out fires.

Ansible and Red Hat
Ansible was released in 2012 by Michael DeHaan (@laserllama⁵ on Twitter), a developer who has been working with configuration management and infrastructure orchestration in one form or another for many years. Through his work with Puppet Labs and Red Hat (where he worked on Cobbler⁶, a configuration management tool, Func, a tool for communicating commands to remote servers, and some other projects⁷), he experienced the trials and tribulations of many different organizations and individual sysadmins on their quest to simplify and automate their infrastructure management operations.