Automation against the COVID-19 crisis: 4 suggestions to get started
Article by Alessandro Perilli, Senior Director, Management Strategy, Red Hat
Without public cloud computing, we wouldn't be able to face the pandemic in the way we are. On-premise data centers have never scaled this fast, and not even the most rigorous capacity planning in the world would have forecasted the resource consumption we face today. News outlets covering the outbreaks would have not been able to cope with an entire planet constantly refreshing the home page in the hope of reading good news (that’s what I do). Hospitals and research facilities publishing dashboards full of virus spread statistics would not have been able to acquire the massive datasets they have as fast as they did. Videoconferencing and streaming platforms wouldn’t be able to serve, exceptionally so far, the enormous amount of the human workforce suddenly forced to work from home.
And what is public cloud computing in the end? An astonishing, unprecedented, disciplined, methodical, pervasive amount of automation (and a few other, equally critical things).
Automation doesn’t just allow us to cope with the urgency and scale of the demand in the public cloud and inside our data centers. Automation is helping organizations around the world to transition to a work-from-home productivity model. Without automation, the security teams would be hard pressed to install VPN clients across millions of laptops, tablets and smartphones all around the world.
And it’s not just the automation of the digital world. Those items that we still receive through Amazon, Ocado, Walmart and more all around the globe while in self-isolation? Without semi-automated warehouses and distribution centers, no provider in the world would have been able to deliver packages at the speed we are receiving them after weeks of lockdown.
If you need to scale, if you need to cope with the urgency of COVID-19, automation can help. So, what can your organization do to automate now, more and faster?
- Focus on many small tasks rather than a big complex process.
The current situation requires a return on the investment faster than ever. And you cannot realize a return on the investment very quickly if you focus on one big process with many complex subprocesses that have never been standardized or automated before. It’s a waste of resources, and I have seen this approach fail many times in my career. Rather, automate as many small tasks as you can. Aggregated, they represent a significant overload for the company and a distraction from the bigger projects.
The more small tasks you automate, the more you will gain experience with and confidence in the automation solution you selected. At the same time, you’ll have a foundation of automated processes that can become the building blocks of more complex automation projects.
- Review what others have automated.
If you have limited experience, the knowledge of your industry peers will prove invaluable. Find online marketplaces for automation (Ansible Galaxy is an example, but there are others, depending on your automation platform of choice) and study what tasks people automate the most, and how. Evaluate how much is applicable to your IT environment and how much you’d have to change to adapt the automation workflow to it.
- Treat automation like software.
Some automation solutions (like Ansible) adopt a language that is much easier to write, understand, and troubleshoot than actual development code. The easier to understand, the more people can use it in their respective domains of expertise, the faster the return on investment.
Yet, no matter how easy to understand, an automation language is still prone to human error. A risk that you can mitigate by applying some of the best practices in software development. Start by thinking about things like automation workflow reviews or version control. The more mission-critical the environment being automated, the more robust the development process should be.
- Think out of the box.
IT automation is usually associated with the provisioning and configuration of servers in a data centre. While existing automation solutions excel in those tasks, some of them have expanded beyond the realms of IT Operations and are becoming invaluable tools for Network Operations, Security Analysts and Security Operations.
Automation can help configure hardware network devices as fast as they are deployed.
Automation can help deploy new security solutions where you need additional.
Automation can help speed up the triage of security attacks as more bad actors attempt to exploit the explosion of people and organizations going online.
In this time of unprecedented challenge, open source communities around the world are helping fight COVID-19 in many ways: projecting hospital workload, devising medical shields, 3D-printing ventilators, all the way to building high-throughput full testing labs that can analyse up to 10,000 tests per day.
Open source communities also make exceptional automation. Ansible, for example, is completely open source and is one of the top 10 most contributed projects in GitHub, among more than 100 million projects.
That enormous community is ready to help you get started.
