The unconventional guide to the post-virtual machine and post-container paradigm

Ok, so what is serverless?

This question puzzled me for a long while. When I started learning about it, the first lie I’ve discovered was that, unlike what is disclosed by its name, serverless is not about running an application without a server. It’s rather about taking the server out of your concerns as the application owner.

Over the past years, the software industry has come out with a lot of innovation to optimize the management of the underlying physical machines that our code runs on: virtual machines, cloud computing, containers, and … serverless.

This latest paradigm is the fastest-growing cloud service in the market. CB Insights has pointed out that  serverless is expected to reach $7.7B by 2021, up from $1.9B in 2016.

It is worth putting some precise explanation around this model so that we can at least clarify our understanding of what we’re getting. In this article, I will share some enlightenment on this topic.

To start, I’ll drill into a bit of terminology, and then I’ll explore the benefits and drawbacks of this technology. I’ll tackle as well what differentiates it from other cloud computing approaches by answering some key questions around the subject.

Table of Contents

Lingo
  ∘ Autoscaling
  ∘ Load balancing
  ∘ Backend as a Service (BaaS)
  ∘ Function as a Service (FaaS)
  ∘ Per-request pricing model
  ∘ Serverless architecture
  ∘ Startup latency
  ∘ Cold start
  ∘ Warm start
  ∘ Vendor lock-in
1\. History of Software Deployment
2\. Does Serverless Replace Previous Deployment Models?
3\. Do Serverless Providers Take Only Care of Servers?
4\. Serverless vs. Containers
5\. Serverless vs. PaaS
6\. Does Serverless Take DevOps Out of Our Concerns?
7\. How to Build a Serverless Application?
8\. Is a Microservices Architecture Required for Serverless?
9\. Can I Run a Monolith Application as Serverless?
10\. What Are the Limitations of Serverless?
Takeaway

#web-development #serverless #aws