Car factory assembly line

How to take an architectural approach to IT automation

Human error affects IT teams in virtually every industry. High-profile errors are being caused by a combination of factors facing IT today: insignificant budget and greater demand.

A recent incident underscores how severe the consequences of human error can be. Amazon Web Services’ Simple Storage Service (S3) suffered an outage that disrupted service for many of its customers, including high-profile sites such as Netflix, Spotify, and Expedia. While S3 came back online in a few hours, the response time was not adequate; the outage lasted long enough time to cause havoc. In fact, 54 of the top 100 Internet retailers were affected by at least a 20% decrease in performance, said the web monitoring company Apica.

Once services were back up and running, attention quickly switched to the cause of the outage. It turned out to be human error—a simple typo. That's all it took to cripple a large section of the Internet. 

Such incidents illustrate that it's time to update our thinking on IT automation. Here's what a modern architectural approach to automation should look like.

The best cloud and IT Ops conferences of 2017

IT automation: Doing more with less

This is where IT automation comes into play, helping organizations do more with less. By reducing time spent on repetitive tasks, the risk of error is minimized and there is more time for IT to focus on mission-critical work. IT automation also adds a system of checks and balances by preventing unauthorized changes. And if a mistake is made, automation provides the ability to quickly roll back the change made in error. 

However, the part that is often overlooked is how and when automation is integrated into enterprise IT. Traditionally, organizations have taken an elemental or piecemeal approach to automation—deploying point solutions to solve point problems. While this elemental approach may work in the short term, not taking a more universal approach means having disparate solutions that don’t communicate, reducing end-to-end workflow efficiencies.

The answer? An architectural approach to automation. This puts processes and systems first by consolidating and coordinating silos of automation within a single framework. This gives businesses the IT agility they need and reduces the time expended on managing and supporting multiple solutions. It also means ensuring that automation is researched at the start of a new process, not as an afterthought. While this may sound simple enough, in order to successfully transform IT processes from an elemental approach to an architectural approach takes some legwork up front and requires a clear channel of communication between all departments within an organization. In the end though, adopting the new way of thinking will have noticeable and lasting ROI.

Setting the groundwork

To achieve enterprise-wide control, visibility, and management of automation processes, key questions need to be asked, and steps need to be taken. Here are a few examples:

  • Take an inventory of all automation solutions across the entire company. The dynamic needs of today’s businesses are driving organizations to increasingly adopt new technologies and applications that often don’t play nicely with the existing set of infrastructure and legacy systems in place. The only way to have a clear picture of how the automation process can be improved is to understand what you are working with. Look for redundancies, as well as disparate tools, applications, or servers that aren’t properly integrated.
  • Have all departments provide insight into current processes that are done manually. It’s surprising how many processes are done manually because that is simply the way they have always been done. However, less manual intervention means a major reduction in time and cost, as well as increased reliability and improved rates of compliance with service-level agreements.
  • Identify when automation is integrated into a process. Is automation integrated at the beginning of the process, or considered after it's been completed? Ideally, automation decisions should be made in the design and conception part of the project, and not left as an afterthought.
  • Determine how processes are currently automated and how these processes are managed. Scripting—a time-intensive process that is prone to errors—is traditionally how IT connects heterogeneous applications, databases, and platforms that were not designed to work together. While scripting won’t, and shouldn’t, completely go away, organizations need a way to bring together business processes in a central location, with or without custom scripts. With the right IT automation system in place, it’s possible to manage scripts in a centralized location, access revision history, implement version control, and utilize granular scheduling capabilities. As a result, IT organizations can minimize the risk of business disruptions and achieve a more fluid technology delivery to the business.

The way forward on IT automation

Mistakes happen all the time; some just happen to play out on a bigger stage than others. No IT system or process is perfect, be it Amazon’s S3, whose outage caused hours of frustration for some of the best-known technology companies, or the US Customs and Border Protection processing systems, which were taken down by a computer glitch, stranding thousands of travelers. However, forward-thinking companies are adopting technology and solutions, such as IT automation, to help mitigate mistakes caused by human error, antiquated IT systems, and more.

While it can be tempting for companies to adopt and deploy the latest and greatest technology solutions with only the immediate benefits in mind, IT automation should be considered as a prologue, not a postscript, in IT planning. This means flipping traditional thinking on its head—eliminating the time-intensive task of first creating processes manually, producing the output needed, and then thinking of automation. This change in thinking is key when considering how to maximize and IT automation strategy.

The best cloud and IT Ops conferences of 2017

Image credit: Flickr

Topics: IT Ops