Back to all blogpostsServerless technologies are an emerging trend that has the potential to revolutionize the way we architect and build applications in a faster, more efficient manner. Serverless frameworks provide developers with the ability to create robust enterprise-level applications without having to worry about servers or infrastructure. This blog post will discuss Serverless use cases and examples, as well as best practices for using Serverless frameworks in your next project!
Why businesses leverage Serverless
Serverless architectures are a potential game-changer for organizations that need to scale their applications quickly and efficiently. Serverless frameworks allow developers, system administrators, and DevOps teams to focus on solving business problems rather than maintaining servers or infrastructure.
Serverless technologies include AWS Lambda functions as well as Cloud Functions from Google Cloud Platform (GCP), Azure Functions, IBM OpenWhisk Serverless Framework – these all help businesses with the following:
- Rapidly prototype new ideas without long lead time investments.
- Support legacy codebases alongside newer serverless backends.
- Deploy scalable microservices architecture at low cost by only paying per use of computing resources.
- Manage costs more effectively by using pay-per-use pricing models.
- Develop applications on diverse platforms and languages with Serverless frameworks.
As the volume of demand increases and decreases, serverless computing and architecture generally scale up and down on autopilot. Additionally, this solution may significantly reduce operational costs and server maintenance.
How companies scale their business and serve their customers by going serverless
Serverless framework use cases span a variety of industries, from fintech to healthcare providers. Serverless architectures are also being used in emerging fields such as the Internet of Things (IoT), artificial intelligence (AI), machine learning, and blockchain technologies – just to name a few!
Here are a few companies that stand out with their serverless offerings.
Netflix going serverless
It goes without saying that Netflix is one of the biggest fish in the broadcasting market, serving millions of customers all around the world. Of course, it requires a huge backend infrastructure to carry a seamless customer experience.
Amazon Web Services Lambda is used by Netflix to run tasks that would otherwise require a great deal of computing time and effort.
During the AWS re:Invent event in 2014, Neil Hunt, Chief Product Officer, has explained the change AWS Lambda brought to their existing infrastructure management.
As of now, Netflix’s media encoding process is automated by event-based triggers. As well as backing up and deploying instances at scale, they made it easy to validate the completion. Each infrastructure process falls into place thanks to the continuous monitoring of AWS resources.
AWS Lambda makes publishing easier as well. Netflix receives thousands of files every day from publishers, and each of those files must be encoded and sorted before delivery to the user. AWS Firebase triggers an event once the files have been uploaded to S3 in order to call the Lambda function that splits the video into 5-minute parts to be encoded into 60 different streams required by Netflix. Using a series of rules and events, the final part of the video is aggregated and deployed.
AWS Lambda is capable of alerting and shutting down instances if unauthorized access is detected as well. Moreover, Netflix can easily pinpoint the cause of an issue when something goes wrong, given that Lambda routinely verifies that the files are accurate, valid, and backed up.
Read more on how Netflix completed the migration.
Zalora going serverless
Asia’s largest fashion retailer, Zalora, features a collection of almost 8,000 products and operates in 8 countries of Australasia. Thus, as you can imagine, there are quite a few orders to be processed. Serverless comes into play here.
Zalora heavily relies on serverless technology to serve millions of orders every single month.
| “We outgrew what the server world offered to us and it was the right time for us to switch over to a provider like AWS” |
The company uses AWS to ensure that its customers have a pleasant experience when shopping on the site. The same applies to the mobile app that also runs on serverless technology. Using AWS and Lambda, Zalora doesn’t have any problems with the scalability of the number of requests they receive, and Zalora can offer customers faster access to features using a serverless framework. Zalora has also been able to control costs by scaling with AWS. Starting from the smallest SAP HANA-certified instance size 244 GB, Zalora could scale the Amazon Elastic Compute Cloud (Amazon EC2) instances to meet growing customer demands. Zalora’s cloud consumption on Amazon Web Services increased nearly 900 percent in two years.
AWS Lambda and AWS Step Functions, in addition to serverless computing features, are also used by Zalora to automate operational tasks. Therefore, the SAP refresh time has been reduced from five days to two days, and the refreshing quality has been improved. As a full-service, centralized backup system, AWS Backup is available to Zalora. AWS is believed to have increased Zalora’s product delivery time by 3 times.
Figma going serverless
Figma simplifies the process of creating professional designs for people. A design platform based on open standards developed in 2012, Figma competes with Adobe Photoshop in offering an alternative.
With millions of clients and projects all over the world, it goes without saying that it needs an infrastructure that stands out. The use of serverless architecture allows designers, developers, and marketers to focus on design and collaboration rather than managing files.
Figma also allows multiple people to work simultaneously on one task and project, and of course, that’s caused by the right implementation of serverless technologies. It works in the cloud, without any installation, making it a seamless experience for users.
Nordstrom going serverless
Nordstrom is a major department store chain based in Seattle, Washington. The company has always been on the front lines of technology innovation.
Rather than using data-driven applications, they switched to event-driven applications and created an open-source Serverless architecture retail store called Hello Retail as a proof-of-concept.
Nordstrom created an event-driven app with the help of AWS Lambda. The Hello, Retail! experiment by Nordstrom demonstrates how they use Kinesis and several other AWS services. Serverless, event-based, and based on an immutable, ordered, and distributed ledger, this project is made up entirely of serverless components.
Coca-Cola going serverless
The Coca-Cola Company has been migrating to serverless technology since 2016 and they are still moving everything in the cloud, but since they still have old technologies in use, traditional servers are being used until they can decommission them. The Coca-Cola CEO Michael Connor spoke at the AWS Re-invent conference about a few case studies they did on serverless in which they saw a 65% cost reduction by using it.
Coca-Cola vending machines located around the world are linked to Coca-Cola headquarters through an integrated communication system.
What Coca-Cola did before going serverless?
They have about decade-old vending machines.
When the client purchases a drink, the machine contacts the payment gateway to confirm the purchase, making the API rest call to use the AWS API Gateway, which triggers a lambda response. The transaction will be handled entirely by the AWS Lambda service. In the case of mobile transactions, a fifth step is required, which is a push notification to a user’s phone that submits the information to Android Pay or Apple Pay.
Until 2016, they have been using 6 EC2 T2.Medium machines cost them almost $13k annually to run. Adding up all the costs for all the features they needed, it came to $4,490/year after moving to a serverless framework.
This roughly means $8,500 savings per one machine per year. With an estimated, whopping 3 million vending machines around the world, this brings more than 25 million dollars in savings.
The Coca-Cola Company has taken serverless to such a high degree that they have made it so that whenever you have an idea for a project to be considered by the architecture review board, the idea needs to be based on serverless. The others are basically rejected.
T-Mobile going serverless
Similar to Coca-Cola, T-Mobile has adopted a “serverless first” policy for any new services it develops. In fact, their experiment with AWS serverless architecture resulted in a reorganization of their whole company.
| “Jazz and serverless computing on AWS have cut out so many time-consuming development and deployment steps. That’s time we can spend innovating and iterating on the solutions customers want to see. Jazz and serverless computing on AWS are major accelerators in how quickly we can give our customers what they want.” |
In order to work with cloud-native APIs, functions, and websites on serverless, T-Mobile developed Jazz. Test, automation, security, architecture, and integration are just some of T-Mobile’s requirements before deploying applications. By using a reusable code template and an easy intake process, Jazz simplifies these tasks. The developers can innovate more rapidly and provide a better customer experience when they use a serverless framework, as they save a lot of time. The T-Mobile team claims they have also saved money since switching from paying for actual resources to only paying for what is actually used, another great benefit of serverless that each modern business should definitely consider.
By offering fully customizable templates, admins are able to define new ones easily, and developers can expose them as services. Developers can build serverless applications using Jazz’s rich ecosystem that supports Serverless Framework.
You can even view it on Github.
T-Mobile uncovered a number of benefits from using AWS serverless technologies for its APIs, microservices, and time- and event-based processing, including smoother scaling, improved resources, reduced patching, and time savings. Increasing flexibility in responding to customer demands was made possible by an even stronger focus on exploration and innovation.
Over to you
Are those the only companies that benefit from a serverless approach? Definitely not! We are sure that more companies are going to join the list in the future.
