The benefits of Kubernetes (K8s) are widely touted in the tech community. It can reduce your hosting costs by increasing efficiency. It’s self-healing and can rollback to recover from errors automatically. It lets development teams ship faster so that companies can innovate and get to market quicker. It can walk the dog for you… Okay, well maybe not that last one.
The point is, there’s a lot of talk about how this container management tool can simplify devops and act as a tool for digital transformation and business growth. We covered this in more depth in our Beginner’s Guide to Kubernetes. But where’s the evidence? Adopting Kuberentes is a big step. Any legacy applications will need to be transitioned to microservices and containerised, then there’s the cluster setup and ongoing maintenance and monitoring. Plus the work associated with keeping Kubernetes itself updated. So how can you be certain that K8s really can benefit your company before you commit?
We’ve rounded up some use cases and stats from early adopters who’ve fed back on their experiences with Kubernetes. They span various business types and industries so you can better understand how Kubernetes could help you to meet your business goals.
Company Size: 40,000+
Capital One is a leading retail bank. They have applications that handle transactions, fraud prevention, credit applications and more, with each one processing hundreds of actions every minute of the day. In the financial sector, it is important that they can update these applications quickly and safely. Time is, quite literally, money.
They began their cloud-native journey back in 2016, first adopting Docker and then Kubernetes. Since then, they have seen impressive improvements to their productivity and reductions in operating costs. Here are some of their key stats:
They attribute the increased resilience and workflow speed they’ve seen to the scalability and coordination offered by Kubernetes.
Company Size: 3,000+
The Financial Times (FT) is an international broadsheet newspaper publishing articles online and in-print daily. They were early adopters of containerisation back in 2016. After one year they had over 150 microservices which had lots of interdependencies and were being managed by multiple teams. It became apparent that they needed a stable, coherent way to manage everything. So in 2017, they decided to migrate to Kubernetes.
Sarah Wells, technical director for operations and reliability, has since commented that the team are now able to focus on work that improves functionality rather than just maintaining the platform. They have also been able to reduce their resource overheads, going from one virtual machine (VM) per service, to just 8 VMs in total. Take a look at some of the improvements they’ve seen:
The huge increase in total releases per year is down to the fact that FT could previously only release approximately once per month, they are now able to make 15 to 20 releases a day!
It is worth noting the production incident figures are actually based on a single month in 2017 (pre-K8s) versus the same month in 2018 (post-K8s migration), rather than the average number of monthly incidents. However, this still demonstrates a year on year reduction in the number of issues in production environments. Plus, the rollback functionality of Kubernetes means that post-migration production incidents were resolved automatically, without the need for developers to intervene!
Industry: IT Services
Company Size: 10+
CloudBoost provides cloud services e.g. storage and auth to help app developers to quickly build powerful and scalable apps for web and mobile.
They shared feedback on their adoption of containers and Kubernetes back in 2017, when they were running a 50 node cluster, and demonstrated great results. Positive impacts were seen on cost, scalability, time to production and customer satisfaction:
Scalability is a huge part of CloudBoost’s business – it’s why their customers use them. If there’s an issue with application scaling then their engineers receive an alert. Since launching auto-scaling on Kubernetes, the need for these alerts has been almost completely eliminated. This, paired with the fact that they can now ship customers code to production in minutes rather than seconds, has improved satisfaction levels.
The engineering team has also benefited from simplified builds. Instead of having to work with a huge single code base, each individual can now really focus on one or two services.
Company Size: 10,000+
Ocado is an online-only supermarket, delivering food and household items to thousands of customers. They developed a single platform (Ocado Smart Platform) to allow them to manage everything from their warehouse to their websites. In recent years, they have expanded into licensing this software to other retailers.
In 2016, they realised that in order to keep scaling they needed all of their services to be accessible through one API to streamline rollouts – enter Kubernetes.
They are now working with approximately 150 microservices in a single cluster. Previously, they were provisioning their services using fleet which required them to use 10 different clusters! Here are some of the other reductions and improvements Ocado have been able to make:
Having everyone working on a single shared cluster has increased visibility and awareness within the development team. This has been a major benefit in their mission to have a more unified platform.
Mike Bryant, a platform engineer, has also commented on the flexibility of Kubernetes. The ability to ship features faster whilst not being forced into working in one particular way has reduced the workload on technical staff and enabled Ocado to better serve their customers.
These case studies demonstrate the business value that can be gained from adopting Kubernetes. If your company is ready to take this digital transformation step but aren’t sure you have the time/resources/first clue where to begin, Bytemark can help!
Complete the form below to register your interest and we will be in touch to discuss your needs:
We offer an end-to-end Kubernetes service. This means that our team of experts can help you at every stage:
- Transition applications to microservices.
- Containerise workloads.
- Install Kubernetes.
- Architect and deploy a solution that fits your specific needs.
- Ongoing cluster management and operations support.