Virtualisation is the practice of running event or process simulations on a desktop, or through the cloud. It can be useful in creating models or scenarios for new products or scheduled events.
For IT administrators, virtualisation cuts down on hardware costs and maintenance, but adds a host of new management tasks. Performing each job of monitoring, provisioning or maintenance is not only time-consuming, but leaves room for human error.
Many data centres use automation tools and technologies, to speed up these processes.
Benefits of Automation
1. Physical and virtual servers can serve YOU.
An automation platform can transform your IT infrastructure into code – and it’s this code which will empower your servers. Whether your network is in the cloud, on-site, or a hybrid, an automated system will help you to easily configure, deploy and scale your servers and applications.
2. Savings
Automated infrastructure can accelerate your time to market, and help you manage scale and complexity. With automation, you can set up your infrastructure, ready to deploy new features in minutes, not days.
3. Adaptability
By transforming your infrastructure into code, you can build, rebuild, configure, and scale in real time.
4. Safeguards
An automated system can monitor for exceptions and unplanned events. If disaster strikes, it can even be configured to reconstruct your entire network.
5. Easy Maintenance
Automation can make your infrastructure easier to maintain, giving you reduced downtime, and increased visibility into your operations.
The DevOps Stratagem
DevOps is an umbrella term for anything that smoothes out the interaction between development and operations. It’s a complex dilemma, but the basic argument goes like this:
Development activity tends to come from a mindset where change is the thing that people are paid to accomplish. The organisation depends on them, to respond to changing needs.
Operations tend to come from a mindset where change is the enemy. The business depends on these people to keep the lights, on and deliver the services that will make money, today. Operations activity is motivated to resist change, as this undermines stability and reliability.
Moreover, development and operations teams tend to inhabit different parts of a company’s organisational structure – often in different geographic locations.
Yet, both streams have to work together, if your business is going to succeed. To achieve this, you’ll need:
1. Incentives to Change Old Habits: All parties involved in the development-to-operations cycle need to understand their stake in the larger business process of which they’re a part.
2. Unified Processes: The entire development-to-operations cycle must be viewed as one end-to-end process. Individual methodologies can be used for particular segments of that – so long as those processes can be plugged together to form a unified whole. And each process must be managed from that unified point-of-view.
3. A Common Set of Tools: In the DevOps world, “infrastructure as code”, “model driven automation”, and “continuous deployment” are the watchwords. Which all spell “automation”.
The Tools Available
Among the market leaders in automation platforms, you can choose from:
1. Chef
Available as Enterprise, or Open Source. It will serve to illustrate the basic workings of a platform.
Chef relies on reusable definitions known as cookbooks and recipes, written using the the Ruby programming language. These elements automate common infrastructure tasks. Recipes are step-by-step instructions for assembling ingredients into a complete, running system.
In turn, the cookbooks and recipes are made from building blocks called resources – included in the platform, but you can add your own. There’s also an online community of Chef users, with whom you can trade resources, cookbooks, or recipes.
The Chef server stores your network’s configuration data and recipes. The data describes all the “ingredients” making up your infrastructure.
The Chef client is a program that runs recipes on nodes of the network, which may be physical or virtual servers – either in-house or in the cloud.
You use a workstation to update the Chef server from time to time, as your infrastructure evolves. All the changes are captured using revision control.
Enterprise Chef is built for managing and automating large-scale infrastructure, and includes premium features like multi-tenancy, role-based access control, reporting and support from Chef’s automation experts.
Open source Chef is a free version of Chef server that forms the basis for both platforms.
Other products of interest include:
1. Puppet
2. Microsoft System Centre Orchestrator
Best Practices
- Think Scale, and Relevance: Assess your environment, to determine which processes are most worth automating.
- Keep IT on Board: The fear that automation may render jobs obsolete might discourage adoption. But tools which simplify and secure IT processes could bring the best kind of job security in this sector.
- Be Adaptable: Monitor your IT environment for periodic changes. Whatever your chosen method of automated resource allocation, be sure to allow for unexpected spikes in their use.
- Use Things You Can Program, and Program the Things You Use: With the emergence of “as-a-Service” products, you can balance between the level of customisation you need, and the time and capacity you devote to this kind of system management.
- Delegate: You’re probably less efficient at network and data centre maintenance than a typical IaaS provider – so leave that kind of thing, to them.
