What Is Serverless?

Chapter 5. What Is the Future of Serverless?

In Chapter 1 we learned where serverless solutions came from and why they make sense today in an ecosystem of ever-smaller services running on an ever-growing network. We also learned (in Chapter 2) how the nature of serverless computing changes the way both software architects and programmers need to think about their applications. Serverless makes sense for event-driven, long-running services where stability and scale are more important than speed.

In Chapter 3 we focused on the seven features serverless platforms need to deal with and got a look at some comparisons between the platforms covered in this report. Finally, in Chapter 4 we got a brief look at the top three serverless platforms to see how they stack up on our feature list.

Of course, no matter the platform you ultimately start using, the challenges of distributed computing that motivated these new architectural solutions are the same as they were more than 20 years ago, when the trend of creating smaller, widely distributed components started in earnest. We still need to deal with the network, we still need to figure out how to design and maintain flexible systems, and we still need to implement our solutions in ways that make them not just easy to deploy but also easy to monitor, maintain, and update over time.

Realities of Distributed Computing

We started by talking about L Peter Deutsch’s list of distributed computing fallacies (see “Fallacies of Distributed Computing” ...