The Guardian, Bustle, AOL, T-Mobile, and other companies switched from in-house servers to AWS Lambda over the last several years. For many, it was a monetary decision, whereas others strived to create a better user experience. AWS Lambda is used by companies with established server spaces, and even more so by startups and small businesses that never had powerful in-house storages to begin with because of many serverless advantages.
AWS Lambda is a Function-as-a-Service platform which is a part of Amazon Web Services. It’s a cloud-based platform that allows developers to run code on cloud. The service executes the code automatically, responding to various event-triggers in the application, using the server space of AWS.
AWS Lambda is a popular choice for building serverless apps. Such applications are executed by Function-as-a-Service platforms which provide enough computing space for running functionality but don’t store data. In such solutions, the FaaS provider manages security, scaling, and provides deployment frameworks. While creating a serverless app with Lambda, developers only write code of particular functions and add them to the platform.
AWS Lambda is among the most popular serverless development infrastructures out there. As a part of AWS, it’s integrated with Amazon’s tools for API management, testing, deployment, maintenance, development, AI, and IoT.
Usually, the most common use cases of AWS Lambda for serverless apps are mobile and web backend applications, where Lamba executes server-side operations and business logic while API Gateway directs HTTP requests coming from the front-end side to particular functions.
Another common application is virtual assistants and chatbots because Alexa Skills Kit provides a powerful tech stack for AI development. For IoT, AWS Lambda can be integrated with AWS IoT tools.
Automated scaling: Lambda automatically increases available server space according to the needs of the project.
Backend development: the platform offers custom APIs for integrating new backend functionality and optimizing server-side processes.
Native support of popular tech stacks: developers don’t have to learn new frameworks or languages. The platform supports Python, Go, Java, Node.js, PowerShell, Python, and Ruby. It also supports popular SSR frameworks — for instance, you can deploy Angular Universal apps in serverless using Lambda. If your language is not on the list of natively-supported technology, you can use Runtime API to accommodate your code.
Automated maintenance: AWS provides a deployment environment, built-in usage monitoring and log management, and security patch releases.
Function coordination: AWS provides error-handling, parallel, branching, and sequential algorithms for workflow management. In Lambda, all code is broken down in functions, which can be edited one by one at any given moment.
Lambda software is composed of a number of functions. Creating a serverless app with AWS Lambda is built upon event-based development, so a user’s action serves as a trigger that kickstarts backend and frontend response.
Lambda’s role is to connect users’ actions to the software’s output, making sure that an application smoothly reacts to interactions. It can be a standalone functionality or feature other Amazon tools (like a DynamoDB database) combining into entire serverless architecture development.
In AWS Lambda, serverless software is deployed with Serverless Application Model, also known as SAM. This is a set of rules that describes the algorithms for serverless app deployment. Here’s how to deploy a serverless app to Lambda.
Install the package for your OS. AWS provides detailed instructions on setting up SAM for Windows, macOS, and Linux.
Deploy your application locally and transfer it to AWS by using the SAM command line. Amazon Web Services offer a SAM package setup, available for download that will upload your code artifacts to Amazon S3, and convert the file to a SAM template. This template will be recognized by Lambda and used to deploy your app on AWS servers. For more detailed instructions on SAM configuration, take a look at Amazon’s official guide.
Automation. After you’ve deployed the app to the Lambda environment, you can use CodePipeline to bypass the step of entering SAM commands manually for every next deployment. You can set up commands and choose deployment methodologies (AWS lets you choose between linear and canary ones), and deliver your code to the server automatically.
Publishing a deployed app. To publish applications, you can use the repository. This step allows you to share code with other developers and edit the project collectively. When you create a SAM file and store it on Lambda, it’s private by default. To change the access setting, you need to upload it to the repository and change the file’s status to the public. The project can be made public (open for all Lambda users) or partially public (accessible to a few selected AWS accounts or via invitation).
Also, after you deploy your app, don't forget to consider serverless monitoring apps in order to get a full traction of all processes on the backend and avoid any critical problems in production.
Flexibility: AWS allows paying only for specific features, such as execution time, amount of invocations and traffic used. If your business often experiences season surges of traffic, Lambda will automatically increase available server space — and return it to the initial stage as soon as the surge is over. You don’t have to pay for unnecessary space or risk server unavailability.
Fast performance: AWS has a powerful server network, distributed across hundreds of locations in dozens of countries. The platform runs from one second to 15 minutes, you can set any timeout yourself.
Automated administration: AWS Lambda automatically tracks application performance statistics, controls memory usage in CloudWatch Logs, records increases of CPU consumption, saves test cases and access permissions, and does a lot of other automated maintenance-related tasks.
Many additional tools and rich AWS infrastructure: Lambda is directly connected to API Gateway, DynamoDB, S3. You can manage data, APIs, use AWS automation tools, and metrics. AWS has more than 212 services, including tools for IoT, AI, Big Data development.
Many businesses switch to Lambda to build web and mobile serverless apps with AWS. The reasons for this transition are the need to accommodate bigger workloads, cost reduction, and functionality redesigns. Here’s why leading companies chose to build a serverless app with AWS Lambda, along with case studies of their transition.
The Guardian was struggling with server failures. Using centralized server space was not a safe option — if one part of a system was down, the entire infrastructure was jeopardized. As a result, customers weren’t getting papers they paid for.
The Guardian was using AWS even prior to AWS Lambda serverless app development. Amazon Kinesis and RedShift were used mostly for analytics. So, the team already knew how to work with the platform, and switching to Lambda for serverless development was a logical choice. Using AWS Lambda functions, The Guardian divided their monolith functionality into several manageable workflows. If one was down, the others were not impacted by the issue.
Distributing functionality across multiple serverless systems contributed to faster performance, better user experience, and simplified development process.
Bustle already used a platform-as-a-service and cloud development to run their computing operations, but it was not AWS right away. The team switched to AWS to get more scaling options and error-free performance.
Prior to their transition, Bustle fasted challenges with the monolith architecture of their website which slowed down its scaling rate. To keep the service running, the team had to involve several managers, automation architects, and monitoring specialists. Overseeing computing operations took up a lot of resources and made onboarding of new tech team members difficult.
Bustle started to use Lambda together with Kinesis Data Streams to get real-time automated updates on website performance. Eventually, they decided to switch to Lambda and transition their platform to a serverless model. Now developers deploy new updates and integrations directly in AWS using Lambda and Amazon API Gateway. The serverless model is used for the web version as well as for Android and iOS apps.
The website for Vogue Italy is an online gallery where photographers all over the world can submit their photos and publish them online. The library stores more than 400,000 photos in high resolution, so an average photo story can be up to 40-50 MB. The number of submitted images grows every day, and at a certain point Vogue servers were not keeping up with the workload.
Amazon Web Services was a logical choice for Vogue due to its rich infrastructure and a flexible pay-per-use system. The tech team transferred its site to a serverless model in less than three months. Now, images are stored in Amazon S3. To process and store photos, the team set up Lambda triggers — user actions lead to automated photoconversion. Event-based Lambda Functions made photo editing and approval a lot easier for the team.
For AOL, the main reason to switch to Lambda was the increase in the workload. Before the transition, the company had more than 14,000 in-house servers. By switching to clouds, the company saved storage pace, energy expenses, and cut their server-supporting staff.
With AWS Lambda, AOL switched its technology products to a serverless architecture. It also uses Amazon Elastic Cloud for data storage and processing and applies a variety of analytical tools. So, even a company that had a robust server architecture to begin with, decided to switch to Lambda due to its speed and cost-efficiency — even though it might not seem like the most intuitive choice at first.
If you are thinking about whether you should build a serverless app with AWS Lambda or transfer an existing platform to Amazon Web Services, you can contact our AWS team. We will help you to set up all the necessary AWS tools — AWS Lambda, Amazon API Gateway, Amazon S3.
Our team helps businesses to transfer their web and mobile projects to a serverless architecture, find the most cost-efficient combination of AWS tools and settings, and monitor the transition process. To get started with efficient, automated, and flexible serverless development, contact our experts — we’ll provide you with a detailed estimate of the project and share our AWS development experience.