A recent report based on analysis of 1.5 billion containers running in IT environments monitored by Datadog finds about 45% of organizations supported by the provider of IT monitoring services are now running Kubernetes in some form. That percentage represents a 10-point gain in Kubernetes adoption over the previous 12 months, the analysis finds.
At the same time, the Datadog report notes that container churn rates are approximately 10 times higher in orchestrated environments. Churn rates in container environments that lack an orchestration platform such as Kubernetes have increased in the last year as well. The average container lifespan at a typical company running infrastructure without orchestration is about two days, down from about six days in mid-2018. In 19% of those environments not running orchestration, the average container lifetime exceeded 30 days. That compares to only 3% of organizations running containers longer than 30 days in Kubernetes environments, according to the report’s findings.
The most widely employed instance of Kubernetes as of October was Kubernetes 1.13, which was released in December 2018. At the time of the analysis, version 1.13 of Kubernetes was still the default for Google and Microsoft, while Amazon Web Services (AWS) had moved to version 1.14 by default.
Michael Gerstenhaber, director of product management for Datadog, says it’s clear that there are still a lot of “snowflakes” in that most organizations have not been able to deploy and manage Kubernetes clusters at scale, due primarily to a lack of expertise and tooling. Organizations that have deployed Kubernetes at scale are typically making use of managed services provided by a cloud service provider, he notes.
The degree to which that will remain the case once IT organizations gain Kubernetes tools and expertise remains to be seen. However, Gerstenhaber says Datadog expects the number of IT organizations managing Kubernetes clusters on their own to increase steadily.
In the meantime, the Datadog report also notes Kubernetes usage on Microsoft Azure has grown to the point where it is now gaining in popularity on Azure as much as it is on Google Cloud Platform (GCP). Roughly 80% of Datadog customers running containers in Azure are now using Kubernetes, up from 67% a year earlier. On GCP, more than 90% of organizations running Kubernetes have adopted the managed Google Kubernetes Engine (GKE) service.
Among organizations running containers in AWS, the popularity of Kubernetes has climbed to 45% of Datadog customers. Among companies running containers in AWS, 19% now use Fargate, a managed service for deploying containers, up from 5% a year earlier.
Finally, the report finds 57% of organizations running containerized workloads use Node.js, followed by Java (49%) and Python (36%) and that open source NGINX proxy software runs in approximately 70% of Kubernetes environments. The high penetration of NGINX in Kubernetes environments is likely due in part to NGINX’s popularity as a Kubernetes ingress controller for routing external traffic to services within the cluster, according to the report.
It’s still early days in terms of Kubernetes adoption. However, the report makes it clear that Kubernetes usage is increasing at a surprising rate.