Choose your language:



Hong Kong




New Zealand




United Kingdom

United States

Expert advice on how to choose the right platform as a service for containerization

Building your containerization house with CaaS, FaaS and PaaS

March 31, 2021 | By: Steve Williams

CaaS, FaaS, and PaaS professional surounded by multiple screens dilled with data graphics and stats

Bold organizations understand that speed, sustainability and innovation are enabled at scale by the cloud. If you’re leveraging the power of cloud-based platforms, it’s likely that your organization is headed toward an application containerization strategy. Companies on a cloud computing journey are able to seize inherent containerization benefits through this type of architecture, such as portability and rapid enhancement delivery. In fact, Gartner expects that up to 15% of enterprise applications will run in a container environment by 2024, up from less than 5% in 2020.

But once you’ve made the decision to go the containerization route, there are critical choices to be made. One of the most important is choosing your underlying foundational technology—a platform as a service (PaaS)—to support containerization.

Understand that groups within your organization have different views of containerization

Your organization probably has a variety of teams, each with different goals and motivations that translate to unique perspectives on containerization.

  1. Development teams typically don’t have a vested interest in an underlying technology, as long as the features support what they are trying to do. Development teams want to be able to write modular pieces of code and execute and test it on an ad hoc basis, such as using a function as a service (FaaS) solution. At the same time, developers want to test integrated code through a container as a service (CaaS) solution, and use a PaaS to simplify the build, run and testing of distributed microservices. A flexible PaaS helps speed development of code by automating and simplifying processes associated with code execution and testing.
  2. Operations teams want to spin up containers and keep the services running. They are most interested in specific CaaS functionalities, backed by a PaaS solution. Similarly, when it comes to function-based events, an operations team is most concerned with consistency, which an appropriate PaaS provides. The location of the function is a secondary consideration.
  3. For IT and infrastructure teams, the focus is on diversification across multiple clouds, operating in hybrid mode, or both. They need a PaaS solution that delivers the same deployment interface experience, regardless of whether it’s running in multiple geographical locations or multiple clouds. They don't want to have to rework code based on specific on-premises or public cloud requirements. For example, to enable pipelining, your infrastructure teams don’t want to be pigeonholed into using Amazon Web Services (AWS), Google Cloud Platform (GCP) or Microsoft Azure toolsets. Infrastructure teams are looking for a platform for CaaS that’s scalable, diversified and highly available.

Consider balancing priorities across the organization to realize the most value from your application containerization strategy

As the owner of an overarching perspective for an organization, a decision-maker needs to serve the needs of a variety of stakeholders. For example, an operations team needs to support containers for runtime purposes—they want a CaaS solution that is highly reliable and scalable. Simultaneously, a CTO is more focused on PaaS, as they manage the infrastructure that provides CaaS for application teams. A CTO also needs to support aspects such as availability, performance and geo-distribution on the containers, functions or both.

Leaders need to be responsive and bring products to market, and they are naturally influenced by the teams that they lead. When it comes to containerization, it’s easy to be swayed by a development team, who above all wants to kick off code and rapidly execute. Developers may be attracted to a managed solution like Amazon EKS, which is easy to use and initially inexpensive.

However, some managed solutions have limitations that may not get you to production, such as compliance issues and regulatory considerations. Additionally, some solutions can constrain the business and create extra overhead when the needs of the business necessitate hybrid or multicloud environments. In these cases, a PaaS solution deployed across multiple clouds, on-premises or at the edge greatly simplifies the job of infrastructure teams and provides a consistent interface for DevOps teams. A common PaaS solution also simplifies the support experience, as the underlying infrastructure is handled by PaaS and customized infrastructure management is avoided.

The containerization foundation is the key to creating a modernization roadmap that takes your entire cloud ecosystem into consideration

When building a house, the foundation is one of the most consequential decisions you can make, as everything is built on top of it. The same bottom-up analogy applies to effective containerization strategies.

Underlying foundational technology choices are critical—in this case, the best PaaS.

There are lots of PaaS options that address many different needs. Getting to the right solution requires careful consideration of underlying business challenges and needs across multiple teams and stakeholders, such as support for:

  • On-premises repositories
  • Regulatory requirements
  • Latency requirements
  • Local geography separation
  • Global distribution ingress
  • Common interfaces to simplify development and operations efforts
  • The ability for developers to trigger a function without worrying about the back end
  • Automated scaling for pods and nodes

How to choose the right PaaS for the best cloud experience will ultimately depend on your organization’s specific needs and cloud ecosystem. But by carefully examining those factors, you’ll navigate through the infrastructure transformation with confidence.

Steve Williams is a cloud solutions architect with TEKsystems. He has deep experience in open-source solutions such as Kubernetes and Linux and has worked on infrastructure projects in industries including finance, automotive, manufacturing, healthcare and telecommunications.