Docker Usage Stats: Adoption Up in the Enterprise and Production

There was a flurry of activity around Docker in recent weeks surrounding DockerCon. Some interesting industry reports came out with statistics on Docker adaption. No surprise, Docker usage keeps booming, and when reading the fine print in these reports, some interesting facts come up. For example, the stats regarding enterprise adoption and maturity of the technology for production use. In this blog post we collect the most recently published data regarding Docker adoption, and summarize the main findings.

Note: The data and images used are property of their respective owners, with links to the source materials. The goal of this blog post is to bring together some recent data and identify some common trends.

During the DockerCon keynote of Docker’s CEO Ben Golub the incredible growth of Docker already became clear. Some highlights from his presentation:

• There are 460,000 Dockerized applications, a 3100% growth over two years.
• Over 4 billion containers have been pulled so far.
• Docker is supported by a large and fast growing community of contributors and users
• As an example, there are 125,000 Docker Meetup members worldwide. That is about 40% of the population of Iceland! (Yes, that is the country that beat England at Euro 2016).

A second interesting data point is Datadog’s report on Docker adoption among their users. Datadog is mostly used for infrastructure monitoring, so most of this data is coming from production use cases. Some interesting tidbits:

• 30% increase in Docker adoption in one year.
• Docker runs on 10% of the hosts, up from 2% compared to 18 months ago.
• Docker is mostly used by large companies with a large number of hosts.
• The number of containers running in production quintuples (= 5x) 9 months after initial deployment.

The numbers here might look less stellar than Docker’s own findings, but we we have to remember that Datadog’s customers are mostly (legacy) operational environments, and Docker has only recently seen significant growth for production use cases. Most of the data coming from Docker’s own research is believed to come from non-production environments.

The third, somewhat less elaborate dataset comes from New Relic:

• Average number of containers running at a time per company grew 192%.
• The average container lifespan is decreasing from 13 hours to 9.25 hours with an increased deployment of Docker for very short-lived containers (less than one minute)
• PHP, Ruby, Java, and Node are the main programming frameworks used in containers

Another dataset comes from Redmonk and Bitnami and deals with container usage and orchestration trends. There are also some interesting findings here:

• About half of the respondents use or plan to use containers
• The numbers per company size seem to indicate significant interest at larger corporations, although it is hard to tell since we don’t know the reference size of each bucket
• About 30% of container deployments are in production environments
• Kubernetes remains the leading orchestration tool, closely followed by Docker Swarm, which has a higher adoption than generally believed, certainly in large-scale environments

Finally, the increasing use of Docker in enterprise environments is also confirmed in the RightScale 2016 State of Cloud report (PDF), with a survey of DevOps tools:

• Overall Docker adoption more than doubles to 27% vs. 13% in 2015, and another 35% have plans to use Docker.
• An even higher percentage of enterprises use Docker (29%) and plan to use it (38%).
• 26% have workloads already running in containers: 8% percent in development and 18% in production.
• Container orchestration tools (Kubernetes, Swarm, Mesosphere) are still in the early phases of adoption, but the amount users that plan to use them is more than double than the current amount of users

In conclusion, Docker seems to be keeping its incredible growth momentum, spurred by increasing use in production environments and at large enterprises. This mirrors the trend we see at CoScale, with a significant increase in demand for monitoring Docker production environments during the last 6-9 months.