How containers and DevOps are changing data storage
Containers are fundamentally changing the enterprise. Consider this -- according to 451 Research, the projected market for containers by 2022 is $4.3 billion. And by 2020, Gartner suggests that more than 50% of companies will have adopted container technologies.
While this paradigm shift is dramatically impacting software development, it’s also having a significant impact on infrastructure management, especially storage, where all persistent data is accessed and resides.
Storage for Containers? The Answer is Three Easy Letters: CSI
In the beginning, container deployment was mainly for stateless applications –service components that were only responsible to execute business logic on inputs and pass on the outputs to other services without the need to store information. But as container deployments mature, stateful services like databases are quickly catching up, enabling a complete microservices-based architecture.
Developers tried to bootstrap datastores that they could get their hands on to quickly deploy applications, but very soon it became clear that a standard was needed to interface storage so that containers could evolve into an enterprise-grade workload.
Enter Container Storage Interface (CSI), a standard that allowed storage vendors implement interfaces to the enterprise storage services offered by their storage products to container workloads and do it in an out-of-the-tree model. In fact, infrastructure players like VMware have implemented their value added container storage services like the Cloud Native Storage (CNS) for vSphere based on the CSI standard.
Kubernetes has emerged as the most popular container orchestration platform. But as container deployment went mainstream, setting up and managing Kubernetes clusters for an enterprise scale production environment was something neither the IT Operations nor DevOps engineers wanted to deal with.
Hence the rise of management platforms like OpenShift and Pivotal Kubernetes Service. In fact, at the recent VMworld 2019, VMware announced that they are making significant investments to bring Kubernetes deployment to the vCenter control pane.
What’s more, vAdmins can manage the Kubernetes stack with other virtualized infrastructure side by side in the same vSphere environment and of course with all the software defined datacenter capabilities of VMware. This may have been the best thing to happen to Kubernetes given the reality of a multi-cloud deployment of applications and infrastructure.
Growing Need for Automation
As more applications become containerized, the need for infrastructure players to create easily consumable infrastructure, or infrastructure as code, becomes more critical. This is where customers need to look at their IT Operations and how it is serving the development groups and DevOps processes to see how automation can help make them more relevant in the everchanging application landscape.
On the other hand, infrastructure players need to make it easy to for IT admins to adopt automation tools and enable automated workflows to provision, protect and move data in an application and DevOps-centric fashion.
The rise of containers will also have a huge impact on IT professionals as well. Increasingly, the skills of storage admins and developers are also overlapping. In this new world, storage admins are having to learn to code, while coders must worry about infrastructure.
Given the amount of automation involved, IT Operations specialists need to understand new application architectures, deployment paradigms and train on new technologies and tool sets to lead the change in their organizations. This helps them expand their purview and influence beyond the four walls of the datacenter into the hybrid cloud environment and create new opportunities to grow in their organizations as well.
With the growing number of investments in the space, containers will continue evolving and deployment and management will become easier by the day. But as the space grows, there will be a tipping point in which legacy applications need to be rearchitected to extract the most possible value.
That means that infrastructure and storage admins will play an even more important role in making that a reality. This will create a tipping point where legacy applications will need to be rearchitected. Those IT organizations and individuals that can successfully adjust to new standards, new forms of management and adopt new skills will be the ones that emerge as success stories in this new world.