A study conducted in 2017 estimated that there would be around 2.93 million apps in Apple's App Store by the end of that year. The study also estimated that this number would grow to over 5 million by 2020.
So why are we even mentioning this right now?
Because more applications today are relying on serverless architectures than you think. Even some big brands have started to go serverless.
Don't believe us?
Take a look at these real-world serverless examples. You might be surprised to find out which giants are relying on this technology.
Marbot the Serverless Slack Bot
Slack is a cloud-based platform that provides teams with tools they can use to collaborate. It has become a popular business solution for small companies.
The Slack Bots the app relies on are some of the platform's most defining features. And there's one bot, in particular, that is a serverless application:
Slack's marbot.
This bot sends notifications from AWS to development and operations teams by way of Slack. The bot integrates with cloud applications such as:
- CloudWatch Events
- Auto Scaling Notifications
- CloudWatch Alarms
Lambda Expressions plays an essential part in this process. And if you want to know more about Lambda Expressions, you should check out this blog post. It will help you understand why Lambda is so important when building serverless apps.
Audiobooks With Amazon Polly
Amazon has become a giant, and in more than one way. From sporting goods to scented candles, it sells everything buyers could ever want.
And let's not forget about its hold on the audiobook market. As of 2018, its Audible service was the largest seller of narrated books. The service had sold more than 400,000 titles.
Now Amazon is looking to expand its hold on the audiobook with Amazon Polly.
Amazon Polly is one of Amazon's first AI services. This service turns text into realistic speech. And it relies on Amazon Web Services' CloudFormation to do so.
The process of how Amazon Polly accomplishes this task is somewhat complex. That said, you can take a look at how it works on Amazon's official blog.
Message-Driven Apps
Technically speaking, this real-world example is pretty general. There are, after all, several message-driven apps. And there's nothing that says that every single message-driven app uses serverless technology.
But many message-driven apps rely on serverless tech, and for a good reason. Message-driven applications typically have to collect several pieces of information at once.
Advertisement platforms are a great example in this case. When you click on an advertisement, the platform must redirect you to a new web page. But the platform must also simultaneously record the fact that you clicked on that advertisement.
Why?
Because the platform has to charge the advertiser for each click in a PPC campaign.
This simultaneous collection of information is what serverless technology excels at. The technology processes multiple messages in parallel.
Want More Real-World Serverless Examples?
There are several more real-world serverless examples we could provide. But there are way too many for us to list them all here.
That's how popular serverless technology has become.
In any case, feel free to browse our site for other examples. We've discussed other companies' use of serverless technologies in the past (Netflix, Coca-Cola, etc.).
And if you've used serverless technology before, let us know what your experience was in the comments section below. We'd love to get another perspective on the tech.
Top comments (2)
For clients, I (we) generate AWS Lambda as the service access layer for application control and CRUD implementation. For NoSQL scenarios, we directly use MongoDB and for relational database scenarios we access a generated remote data layer to handle the state maintenance required for fast database transactions. Otherwise, due to the stateless nature of Lambda, firing queries to a relational database directly or through an ORM (like Hibernate) is painfully slow. All in all, performance is not an issue in either scenarios.
We will take a look at Dashbird to help with debugging and monitoring function execution/performance.
I too have seen a lot of companies using Lambda for the simple CRUD implementation and they do seem to be a great fit for those "fire and forget" type functions. I'd love to talk more about the way you guys are using Lambda inhouse. Send me a twitter message @johndemian if you think you have the time to talk.