How Chick-fil-A Became a Kubernetes Success Story

Table of Contents

With the ever-increasing popularity of Kubernetes, it shouldn’t come as a shock that certainly major tech brands, such as Google, DirectTV and Spotify, have migrated to experience the speed and flexibility it provides. 

As more industries catch up to these powerhouses in terms of how they approach their IT infrastructure, future-proofing your business means you either implement the latest technology trends or stay locked in the past. One of the biggest, albeit most surprising, Kubernetes success stories is none other than Chick-fil-A, an American fast-food giant, best known for their cult-favorite chicken sandwiches and stellar customer service. 

Why implement Kubernetes in a fast food restaurant?

For a company like Chick-fil-A, creating streamlined and repeatable processes means consistency, customer satisfaction, and, of course, delicious food. Implementing a cloud-based application architecture might not seem like it runs parallel to those goals, but they started with the big picture.

Chick-fil-A is no stranger to tech. Stop by one of their restaurants and you’ll certainly see the bounty of screens and computer accessories. Everything communicates with each other to help orchestrate the lunch rush dance — and every franchise runs similarly to manage that consistent, seamless customer experience.

So why a containerized application structure? Well, to run the kind of Internet of Things (IoT) framework that Chick-fil-A developed means you’ll need to meet a few aggressive goals:

  • A platform that can roll with the punches and evolve as needed
  • Applications with high-availability
  • The ability to horizontally scale infrastructure and development teams

Essentially, they needed centralized command and control with decentralized operational capabilities in order to allow franchise’s local business operations to tolerate network failures. To meet these requirements, Chick-fil-A turned to Kubernetes. But why choose Kubernetes over other orchestration platforms like Docker Swarm?

If it works, it works 

Chick-fil-A works under the ethos that the only thing that matters is how you function under pressure. For the successful fast food empire, lofty ideas and fancy code don’t mean anything if they can’t perform when it’s crunch time.

So why not choose another method to reach their goals? Well, if stores could run efficiently using a single VMware machine or some other clustering tool, they would. Chick-fil-A was looking for a solution, not a gamble. It wasn’t about implementing the latest technology for the sake of being new. It was all about creating a first-class experience for their customers.

The simplest solution is usually the best choice. In this case, the crucible of real-world adoption showed Chick-fil-A that Kubernetes was that choice, if not the only real choice, to manage the kind of system they needed to keep up with demand.

An easy installation process

At this point, you might be wondering how a complicated application orchestration platform like Kubernetes can work with a network of around 2,000 stores. The cloud, Kubernetes, containerized applications; they don’t have anything to do with making waffle fries and lemonade. So how can such a wide network of stores implement a Kubernetes system effectively?

Chick-fil-A uses a combination of Kubernetes and Edge computing, so the hardware to run the clusters is local to each store. In most cases, it’s a small network of three computers with limited hardware. The hardware investment for owner/operators is just around $1,000. The installation process is incredibly simple, too.

In most cases, workers can unbox the devices, get everything hooked up to power and label the appropriate port … that’s about it. These devices automatically self-bootstrap and cluster. This means even a non-technical person without the knowledge of basic Kubernetes architecture can easily install the system or add on additional hardware when necessary.

Enter: Cloud-fil-A

One of the amazing things about Chick-fil-A’s system is its scalability. Combining Edge computing with Kubernetes means local restaurants are creating their own micro private clouds.

These systems connect to the greater network to transfer key data indicators and other information, all while following Kubernetes best practices. But, for the most part, the system works using a cadre of restaurant network switches and a few routers. This means everything is operating on very available LAN, keeping up with the goal for low-latency and high availability apps.

In a way, this entire system exists as a giant cloud. Call it Cloud-fil-A. Instead of running a few giant Kubernetes clusters with thousands of containerized applications, each store runs its own cluster with just a handful of apps. That means Chick-fil-A has around 2,000 operating clusters — with more popping up with every new store opening.

When a new store opens, attaching to the other stores is as easy as opening up the box and plugging in gear.

Keeping things running smoothly = success 

It’s clear why Chick-fil-A chose Kubernetes over other orchestration platforms, and we have a good idea of what that system looks like for each store. Great. But what does it all do?

The goal of all of this is to simplify the restaurant experience. This puts owner/operators and their teams in a position to optimize the other aspects of foodservice, like serving food quickly and keeping customers happy. The idea is this: make smarter kitchen equipment that can collect data to help optimize and build better intelligent systems.

All that data is incredibly useful, especially if you can gain access in real-time. Let’s say you already have some data on the demand for waffle fries across thousands of stores. This is great and all, but the reality is that restaurants are chaotic, and what is true for some stores won’t matter when six buses full of kids from a local soccer tournament come into yours.

With Kubernetes and the Edge computing system, real-time data coming from the point of sale systems can help inform the software working with the fryers about the work-in-progress inventory. That forecast model we mentioned earlier can see micro-adjustment on the fly when this data gets put into the system.

This kind of strategizing can influence so many different aspects of running a business. Is inventory low? With the open ecosystem that Chick-fil-A produced, developing an automatic ordering system that works with suppliers is on the table. You can even implement robotics and AI into the platform to automatically fry chicken nuggets without needing the watchful eye of a dedicated worker.

The applications of this technology based on Kubernetes architecture are truly remarkable. This is the future of fast food, and Chick-fil-A is leading the way.

How your team can easily implement Kubernetes

Chick-fil-A has harnessed the power of Kubernetes and is seeing the fruits of their labor. While this article paints a rosy picture of what Kubernetes can do for a business, it doesn’t quite illustrate the sheer complexity of what it takes to do so. Chick-fil-A has made it easy for owners to install the boxes needed to connect to the system, but, rest assured, many back-end developments and research dollars went into the process too.

Appvia Wayfinder helps businesses get the most out of a Kubernetes ecosystem without the complexity or needing large, expensive teams of Kubernetes experts on staff. If you want to learn more about how we’re changing how businesses approach Kubernetes, contact us today.

About Appvia

Appvia enables businesses to solve complex cloud challenges with products and services that make Kubernetes secure, cost-effective and scalable.

Our founders have worked with Kubernetes in highly regulated, highly secure environments since 2016, contributing heavily to innovative projects such as Kops and fully utilizing Kubernetes ahead of the curve. We’ve mastered Kubernetes, and experienced its complexities, so our customers don’t have to. 

Share this article
Twitter
LinkedIn
Facebook
profile-112x112-crop-1
Katy Ryder
Demand generation manager
I originally hail from Austin, Texas but now call London home. When I’m not crafting content, you might run into me at a coffeeshop (there’s always room for one more latte), hiking through Hampstead Heath or snapping street style photographs.

The podcast that takes a lighthearted look at the who, what, when, where, why, how and OMGs of cloud computing

Related insights

Managing Kubernetes Secrets with HashiCorp Vault vs. Azure Key Vault Keeping secrets secure...
Namespaces are a vital feature of Kubernetes. They allow you to separate uniquely named...
DevOps teams have rapidly adopted Kubernetes as the standard way to deploy and...
Once you start working with Kubernetes, it’s natural to think about how you...
Self-service of cloud resources Kubernetes has been brilliant at delivering an ecosystem for...
Pods, deployments, and services are just some of the concepts that you need to understand in...
Last week I published a blog, “How to spot gaps in your Public Cloud...
Breaking down the core areas that you should be aware of when considering...
5 tips to help you manage more with less Not every manager of...
Public cloud has provided huge benefits in getting infrastructure and services to people...
This is the story of how three Appvia Engineers contributed so much to...
Overview The UK Home Office is a large government organisation, whose projects and...