Serverless Abstraction with Containers Explained

Introduction

With the rapid growth of cloud computing, the “as a service” business model is slowly growing to dominate the field of enterprise IT. XaaS (also known as “anything as a service”) is projected to expand by a staggering annual growth rate of 38 percent between 2016 and 2020. The reasons for the rise of XaaS solutions are simple: in general, they are more flexible, more efficient, more easily accessible, and more cost-effective.

Serverless abstraction and containers are two XaaS cloud computing paradigms that have both become highly popular in recent years. Many articles pit the two concepts against each other, suggesting that businesses are able to use one but not both at the same time.

However, the choice between serverless abstraction and containers is a false dilemma. Both serverless and containers can be used together, enhancing one another and compensating for the other’s shortcomings. In this article, we’ll discuss everything you need to know about serverless abstraction with containers: what it is, what the benefits are, and how you can get started using them within your organization.

serverless

What is Serverless Abstraction?

“Serverless abstraction” is the notion in cloud computing that software can be totally separated from the hardware servers that it runs on. Users can execute an application without having to provision and manage the server where it resides.

There are two main types of serverless abstraction:

  • BaaS (backend as a service): The cloud provider handles the application backend, which concerns “behind the scenes” technical issues such as database management, user authentication, and push notifications for mobile applications.
  • FaaS (function as a service): The cloud provider executes the application’s code in response to a certain event, request, or trigger. The server is powered up when the application needs to run, and powered down once it completes.

The FaaS serverless paradigm is akin to the supply of a utility such as electricity in most modern homes. When you turn on a light or a kitchen appliance, your consumption of electricity increases, and it stops automatically when you flip the switch off again. The amount of the utility is infinite in practice for most use cases, and you pay only for the resources you actually consume.

FaaS is a popular choice for several different use cases. If you have an application that shares only static content, for example, FaaS will ensure that the appropriate resources and infrastructure are provisioned, no matter how much load your server is under. The ETL (extract, transform, load) data management process is another excellent use case for FaaS. Instead of running 24/7/365, your ETL jobs can spin up when you need to move information into your data warehouse, so that you only pay for the run instances that you actually need.

serverless abstraction

What are Containers?

Containers are software “packages” that combine an application’s source code with the libraries, frameworks, dependencies, and settings that are required to use it successfully. This ensures that a software application will always be able to run and behave predictably, no matter in which environment it is executed.

Products such as Docker and Kubernetes have popularized the use of containers among companies of all sizes and industries. 47 percent of IT leaders plan to use containers in a production environment, while another 12 percent already have.

Serverless Abstraction with Containers

The goal of both serverless abstraction and containers is to simplify the development process by removing the need to perform much of the tedious drudgery and technical overhead. Indeed, nothing prevents developers from using both containers and serverless abstraction in the same project.

Developers can make use of a hybrid architecture in which both the serverless and container paradigms complement each other, making up for the other’s shortcomings. For example, developers might build a large, complex application that mainly uses containers, but that transfers responsibility for some of the backend tasks to a serverless cloud computing platform.

In light of this natural relationship, it’s no surprise that there are a growing number of cloud offerings that seek to unite serverless and containers. For example, Google Cloud Run is a cloud computing platform from Google that “brings serverless to containers.”

Google Cloud Run is a fully managed platform that runs and automatically scales stateless containers in the cloud. Each container can be easily invoked with an HTTP request, which means that Google Cloud Run is also a FaaS solution, handling all the common tasks of infrastructure management.

Because Google Cloud Run is still in beta and under active development, it might not be the best choice for organizations who are looking for maximum stability and security. In this case, companies might turn to Google Cloud Run alternatives such as Iron.io.

Iron.io is a serverless platform offering a multi-cloud, Docker-based job processing service. The flagship Iron.io product IronWorker is a task queue solution for running containers at scale. No matter what your IT setup, IronWorker can work with you: from on-premises IT to a shared cloud infrastructure to a public cloud such as AWS or Microsoft Azure.

containers

Conclusion

Although they’re often thought of as opposing alternatives, the launch of Google Cloud Run and alternatives such as Iron.io proves that serverless abstraction and containers can actually work together in harmony. Interested in learning more about which serverless/containers solution is right for your business needs and objectives? Speak with a knowledgeable, experienced technology partner like Iron.io who can help you down the right path.