UPDATED 11:10 EDT / FEBRUARY 21 2012

NEWS

Guest Post: Why Model-Driven Configuration Management Rules

Written by Luke Kanies, Founder and CEO, Puppet Labs

Automating the deployment and configuration of systems is meaningless if the automation isn’t a force multiplier, streamlining the adoption of new technologies and practices while maintaining service levels.

Many organizations without configuration management tools use a suite of scripts written in an interpreted language to deploy and manage their environments. These scripts evolve over years as they’re passed from sysadmin to sysadmin, growing to contain the personal quirks of their developers and a long history of legacy workarounds. They generally start life as point solutions that made some small, critical job easier, but the scripts eventually calcify into an unmaintainable mess that makes it easy to apply current practices but nearly impossible to adapt to changing technology or policies.

Thus, instead of script-driven automation turning operations into a competitive advantage, it halts innovation in IT and forces all work through the few people who understand what the scripts do and how to use them. This kind of gate-keeping and resistance to change is exactly what has led many organizations to look outside the firewall at hosted services and the public cloud, but there is a simpler, better way to convert IT from a tar pit of inaction to a vector for innovation throughout the organization.

Model-driven configuration management provides all of the power and flexibility of script-driven systems while adding unprecedented organizational leverage. The key to model-driven configuration management is that it results in a highly-maintainable system that is always available for analysis and auditing. Even novice users can take a given configuration and understand what result it will have on the system and why that result is important. This ease of access to critical information means that the automation becomes a tool that IT can easily maintain and the rest of the organization can rely on for clear status information and better control of their own needs.

In a script-driven system, the only way to determine what a script will do or whether it will work is to run the script on a test host or to spend enough time to understand the script deeply. This is reasonable on a small network, with a small amount of scripting, and enough time to do the job right, but today’s networks are moving faster, at a larger scale, and with shorter lead times than ever before. Model-driven configuration management produces tremendous amounts of data, which provide many new views into the automation system, so you can understand what the system is doing and how infrastructure changes impact service levels, providing the information needed to pro-actively plan accordingly.

In addition to this greater support for analytics, model-driven management provides a complete framework for logging, error management, and reporting, with no need for the sysadmin to build it all from the ground up – all for free. This allows the sysadmin to focus on building configurations, and thus get more done in less time with a much smaller code base to maintain over time. Even better, when custom tools are required in a model-based system, there are obvious integration points and APIs for translating the model behind the automation into pre-existing systems such as inventory management, change control, and analytics dashboards. These tools allow a larger portion of the organization to take advantage of the automation without any of them having to directly understand how the system itself works.

An important consequence of model-driven configuration management is the availability of control points that allow managed operation without requiring external, manual control processes. In a traditional script-driven system, you only have as much control as the script author has been able to develop, but in a model-driven system, the framework provides control and visibility at multiple points in the process. For example, you can have a technical user reviewing code before it makes it into production while a business reviews changes on individual hosts. These control points are a critical step toward providing confidence that planned changes match user expectations.

The combination of vastly better information and automated processes with cross-organization management means that technology can be adopted with confidence, rather than fear of disruption. Instead of paper processes that slow everything down, you can rely on automation with control points directly exposed to the decision makers. This often results in service deployments being orders of magnitude faster, which means ROI on technology investment is both higher and sooner.

Other tools give sysadmins powerful methods for deploying and triggering scripts or applications remotely, and many contain lots of useful features and are immensely powerful and scalable. Yet without a model-driven system providing leverage throughout the organization, what sysadmins really have is an old tangle of scripts with a few gate-keepers to getting anything done.

Model-driven configuration management will bring system administrators much closer to solidifying a process that is successful, reliable and predictable. It rules. Why would you want to risk running your environment any other way?

Editors Note: This is a guest post by Luke Kanies, Founder and CEO, Puppet Labs

Luke Kanies is the founder and CEO of Puppet Labs and the founder of the Puppet project. Previously, he was a consultant, open source contributor and article author. He has focused on tool development since 2001, developing and publishing multiple simple sysadmin tools and contributing to established products like Cfengine. He has presented on Puppet and other tools around the world, including at OSCON, LISA, Linux.Conf.au, and FOSS.


A message from John Furrier, co-founder of SiliconANGLE:

Your vote of support is important to us and it helps us keep the content FREE.

One click below supports our mission to provide free, deep, and relevant content.  

Join our community on YouTube

Join the community that includes more than 15,000 #CubeAlumni experts, including Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger, and many more luminaries and experts.

“TheCUBE is an important partner to the industry. You guys really are a part of our events and we really appreciate you coming and I know people appreciate the content you create as well” – Andy Jassy

THANK YOU