Key Considerations in Container Adoption

Today, containers have attained near-ubiquitous adoption as an application deployment medium—and with their proven stability and increased standardization, container formats and their runtime engines are becoming a de facto deployment option for much more. Many organizations are running containerized workloads with ease; for instance, it’s not uncommon to see microservices deployed in containers and running on a service mesh architecture.

Let’s take a step back to where it all began. Docker container technology launched in 2013 as an open source Docker Engine and revolutionized the tech industry. It created new opportunities for not only developers, systems operators and the general software engineering community, but also for new companies, organizations and technologies to meet the unprecedented needs of managing containers effectively.

The container community moved quickly to create new technologies that could successfully orchestrate containerized workloads in a complex deployment model. Meet the leader of the pack: Kubernetes. In 2015, Kubernetes came to the forefront of the container industry and became one of the most widely adopted container orchestrators. That has only grown over the past five years; Kubernetes has become the go-to for container orchestration.

Key Considerations for Container Adoption

Containers can provide significant benefits, such as increased resource efficiency, better density, improved isolation from failures—and even better development hygiene—by removing environmental inconsistences of trying to build software on differentiated environments. As containers increasingly become ubiquitous, all people involved with them must understand the nuances of this ephemeral world to avoid painful mistakes in their implementation.

While some organizations have been running containers in production for several years, some aren’t quite there yet. There’s a steep learning curve with orchestrating containers effectively in production and many organizations lack the expertise to navigate issues during container adoption successfully.

With Kubernetes celebrating its fifth anniversary this summer, we wanted to take this opportunity to explore the state of container adoption overall and share some top considerations for managing the growing ubiquity of containers:

  • Prepare for deployment: While containers uncomplicate the environment from an application developer perspective, in many ways they complicate deployment infrastructure—instances exist and then disappear. Because of this, it’s necessary to focus on the overall health of the containers, identify what their interaction patterns are, ascertain service communication schematics and understand where containers work well—and where they don’t. It is also key to understand the concept of identity (service accounts) within the container world; understanding how you will proceed with container deployment will help ensure a successful and pain-free journey.
  • Leverage monitoring tools that span various infrastructures: Regardless of where you may be running containers—on-premises; in the cloud using services such as Amazon Web Services (AWS), Google Cloud Platform (GCP) or Microsoft Azure; or a combination of both—you will need container monitoring tools. Tools that collect and interpret metrics and events across these different landscapes to provide transparent views of infrastructure and application health are paramount to success, in addition to having integrations with open source tools such as Istio directly with container orchestrators.
  • Prioritize security: While consuming open source tools is a key resource for application development, it comes with a host of challenges. For example, without proper management, open source tools may unintentionally introduce vulnerabilities and even malware into your applications. You need to have lightweight, rigid processes and tools in place to respond effectively to such challenges. In the world of containers, services typically are not patched directly; as vulnerabilities emerge, you must rebuild the container with a base image that doesn’t have that vulnerability. It’s important to prioritize security from the start by inserting appropriate security audits on your continuous delivery (CD) process. This includes analyzing code repositories, verifying licenses, performing static code analysis, scanning containers, applying infrastructure and application security policies and more. Securing containers also means securing the application that is running in the container and the process of deploying the containers. As existing application tools and container scanning tools become part of the CD pipeline, you can proactively protect yourself against security vulnerabilities.

Conclusion

As containers increasingly become ubiquitous in agile tech environments and container adoption increases, it’s necessary to understand the nuances of container technologies to ensure you can observe and manage them securely in production. Deploying containers is key to delivering value to the customers quickly and continuously.

Keith Kuchler

Keith Kuchler is the Vice President of Engineering (Software as a Service) at SolarWinds. In his position, Keith strategically leads his team to meet customer needs, critical business objectives, and profit and revenue growth through operational efficiency and innovation. Keith has broad expertise in helping to develop agile and disciplined engineering operations—from startup to broad charter global organizations—and a consistent track record of successful delivery of complex products and solutions. Before assuming the role of VP of Engineering, Keith served as the Senior Director of SolarWinds Engineering, leading engineering for Pingdom, Loggly, AppOptics, Papertrail, Pingdom Server Monitoring, TraceView, and Librato. Previously, Keith served in various roles at Hewlett Packard Enterprise, including as Director of Global Software Engineering: HPE Propel; Engineering Group Manager: HP Propel R&D & Converged Cloud User Experience; and Solution Architect.

Keith Kuchler has 1 posts and counting. See all posts by Keith Kuchler