Containers in context
With the global container management market expected to top $940 million by 2024, up from the $466 million of 2020, organisations would do well to consider integrating these packages of software into their enterprise IT strategies. Of course, there are several components to consider when it comes to the development of a container platform strategy.
One of the challenges facing businesses in this regard is dealing with the myriad of different combinations of software that can be used to manage application containers. This brings an additional layer of complexity to the decision-making process that can slow down the adoption of containers in an organisation, greatly reducing their effectiveness. Some believe that the formula for a container management strategy is the combination of container software plus container orchestration. It all starts with a foundation built on understanding the enterprise requirements of containers and why the company is looking to adopt them in the first place.
Given how a container platform fast-tracks the go-to-market approach of a business, regardless of industry sector, coming to grips with the current IT landscape within the business and what the user constraints have been up to that point are key components to helping define the parameters of what is required. Because containers have been around for some time, many businesses are looking at moving beyond traditional container hosting and trying to identify more sophisticated ways that these solutions can bring value. This becomes more critical given how rapidly hybrid cloud adoption and edge computing have grown since the onset of the COVID-19 pandemic early last year.
Steps in the process
Gartner has identified six elements that should be part of a container platform strategy that can help mitigate the challenges of deploying containers in production environments. These are security and governance; monitoring; storage; networking; container lifecycle management; and container orchestration.
And while each of these are important, it is in container orchestration where the key functionality of such a strategy lies. According to Gartner, the orchestration layer interfaces with the application, keeps the containers running in the desired state, and maintains service-level agreements. Forming part of this orchestration is the scheduling layer that places containers on the most optimal hosts in a cluster. This, in turn, is defined by the requirements of the orchestration layer. Think of this step as the activation and implementation of a container as defined by the enterprise strategy.
It is in the orchestration phase where the lifecycles of containers are managed, controlled, and automated. This encompasses the likes of provisioning and deploying containers; the redundancy and availability of containers; scaling up or removing containers as required; allocation of resources between containers; and the health monitoring of containers and their hosts.
According to Red Hat, container orchestration can be used in any environment where the business uses containers. It can help the organisation to deploy the same application across different environments without needing to redesign it. And microservices in containers make it easier to orchestrate services, including storage, networking, and security. One of the more popular tools to do so is Kubernetes. This open-source container orchestration tool was originally developed and designed by engineers at Google. Google donated the Kubernetes project to the newly formed Cloud Native Computing Foundation in 2015. Kubernetes orchestration enables the organisation to build application services that span multiple containers, schedule containers across a cluster, scale those containers, and manage their health over time.
Side-stepping container challenges
One of the mistakes many organisations make when it comes to containers is to treat them the same as they do virtual machines. And while there are similarities in the functions each of these perform, containers provide new ways of doing software development. Things like stateful and stateless apps, data persistence, and traditional versus cloud-native development processes must all be understood in this regard.
Just like any other IT shift, companies must also avoid vendor lock-in when it comes to the development of their container management strategies. If the strategy is focused on a specific cloud service, it is best to minimise exposure to that service especially when it comes to container portability and the development environments.
Most importantly, companies should not expect their existing operating model to support a container strategy without making significant revisions. Given the sheer volume of changes required by the larger number of containers (or virtualisation units), automation becomes even more fundamental to ease the process. This also requires ongoing reliability checks to be in place to ensure that the automation supports the new operating model.
Ultimately, for any enterprise container management strategy to be successful, an organisation must continually test the container orchestrations. In this way, these can be properly operationalised to deliver on the business mandate and perform the functions correctly while delivering on the uptime and redundancy requirements.
For more information contact us.