Containers are a primary focus for enterprises’ digital transformation strategies today. But, while many CIOs and CTOs have a sense of the advantages that containers could bring, some common misconceptions have formed barriers in their path toward adoption.
We’ve seen the technology take off in a new generation of born-in-the-cloud companies like Netflix and Uber, which acclimatize easily to new IT infrastructures and environments. So, what’s been holding back wider enterprise take up? Some of the biggest misconceptions and barriers that have prevented legacy IT from embracing containers revolve around how they fit into developer workflow and identity management, security, vendor lock-in, and Internet of Things (IoT).
“Containers are a security risk.”
One of the biggest and most common misconceptions presenting an obstacle to enterprise adoption is security. This, in fairness, was a justifiable concern in the early days of cloud containerization. CIOs and CTOs are right to question the security of any new technology they implement; but, the reality is that securing any technology is a grueling undertaking, and there are plenty of controls in place that enable containers to remain secure against cyber threats.
Containers’ greatest strength when it comes to security is isolation. Container technology offers hardware-guaranteed security to ensure that each containerized machine cannot access one another. There may be situations where a virtual machine is required for particularly sensitive data, but for the most part containers deliver security for these scenarios. If one container is infected, its isolation can thwart that infection from spreading, and can be safely shut down if needed.
There are also many tools on the market to keep containers as safe as possible, such as internal scanning to vet the images and detect if anything needs to be addressed.There needs to be a cultural shift for enterprises adopting containers. Developers need to invest in the idea that this technology requires continuous monitoring and updating in order to remain secure. Security is a two-fold effort, and while enterprises should be able to trust the vendors that are creating the containers, IT teams have to take it upon themselves to rebuild the containers every few weeks to remain as up-to-date as possible — minimizing the risk of possible breaches.
With all the measures available, security concerns shouldn’t become a barrier for enterprises adopting containers. However, IT teams should be well aware of the options for keeping them secure before deployment.
“Kubernetes can’t easily attach to my identity management or integrate into my developer workflow.”
For enterprises and developers unfamiliar with containers and container technology like Kubernetes, it can be daunting to think about integrating a new process into the developer workflow. Most dev teams have their own ideas on the most frictionless means of deployment, and are reluctant to change.
Similar to the cultural shift that needs to happen for enterprises to accept containers’ security features, developers must also change their mindset on how Kubernetes, and containers more broadly, is not in itself a full solution. Rather, it’s just part of a solution that needs to be built up and customized to what a developer team wants. Teams will need to take the base pieces that Kubernetes provides and enhances them in a way that works best for them, rather than trying to force the technology as a one-size-fits-all solution.
“We spent years locked in with Virtual Machine (VM). Why would we want to get locked into containers?”
Some c-suite decision makers believe that by using one particular type of container, they will tie themselves to a specific vendor. But, the container ecosystem is far more diverse and open than the legacy VM market. In fact, understanding the correct uses of containers can arm enterprises with the knowledge to choose the right vendor with confidence.
For example, developers and enterprises often see application containers, like Docker, as a way toward better density and performance for existing workloads — which is true. However, to actually migrate an existing workload to an application container requires a lot of re-architecture and work. Machine containers, which boot an OS, provide many of those benefits with virtually no time spent reworking the application.
Different container types are more suited to certain business challenges, and enterprises that aren’t aware of this might write off the technology thinking it won’t help with what they need, when in reality they may just need a different type of container.
“Containers are a cloud-only solution. We’re more concerned with the IoT.”
This misconception treats IoT and cloud as two separate entities. When in fact, the two are intertwined. IoT brings an overload of data, which businesses are struggling to contend with. Yet, the so-called ‘cloud-only’ solution has significant advantages for IoT development, as containers are arguably the ideal response to scalability and data-related issues presented by the predominance of IoT applications.
Containers — Kubernetes specifically — are optimized for enabling the transmission of data between connected devices and the cloud. Containers’ density and the way they’re designed make them easy to scale up and down as needed, which is an advantage when it comes to processing the IoT’s typical large bursts of data.
Upgrading can also be a concern for IoT devices, as they are often intermittently connected and lack the computing power needed to upgrade effectively. Containers can instead help to deliver updates efficiently as new images, put into an IoT device in pre-configured packages, leave all the work to the container and very little to the device.
The number of devices connecting to the cloud won’t soon slow down, and with innovations such as autonomous vehicles on the horizon, there will be a steady increase of devices that need to communicate with the edge and with multiple clouds. As these numbers grow, awareness of the technology becomes more mainstream, and common misconceptions become debunked — containers will emerge as a prominent solution to scaling concerns.