Cloud native computing grows by 200 percent


Over the last few years the way you moved applications from your data center to the cloud was lift-and-shift, refactor, or migrate to containers. The latter has gotten a kick in the pants as cloud-native techniques such as serverless computing and microservices have joined forces with containers.

Still unsure what I’m talking about? Chris Aniszczyk, executive director of the Open Container Initiative (OCI) and the Cloud Native Computing Foundation (CNCF), explained: “Cloud-native computing uses an open-source software stack to deploy applications as microservices, [each part packaged] into its own container, and dynamically orchestrate those containers to optimize resource utilization.”  

You can find proof this methodology is taking off in the latest CNCF survey. This survey of primarily enterprise or DevOps professionals found that “production usage of CNCF projects has grown more than 200 percent on average since December 2017, and evaluation has jumped 372 percent.” Not bad growth for less than a year!

Most of this growth is coming hand-in-glove with the rise of continuous integration/continuous delivery (CI/CD). Sixty-seven  percent of respondents are checking in code multiple times per day compared to 28 percent checking in code a few times a week and 6 percent a few times per month.

This, of course, improves not just how often code is updated, but how often programs are upgraded. Before cloud native computing technologies arrived, release cycles typically only happened once or twice a year. Now, 20 percent of companies are releasing new software versions weekly, with 18 percent monthly, and some brave daredevils, 15 percent, releasing patched programs into production every day.

The plurality, 42 percent, of these high-speed releases are being made automatically. Twenty-seven percent are still doing manual releases, while 25 percent of respondents were using a hybrid release method. As automated releases grow, so does the popularity of CI/CD pipeline tools. Jenkins is the leading tool, 70 percent, followed by Terraform, 27 percent, and the ever popular “custom scripts” at 26 percent.

Simultaneously, serverless technology continues to grow, albeit not at such as an explosive rate. It’s up “only” 22 percent since December 2017. Most respondents are using hosted platforms such as AWS Lambda, 70 percent, Google Cloud Functions, 25 percent, and Azure Functions, 20 percent.

Although those results may make you think cloud native is primarily a public cloud phenomenon, you’d be wrong. The survey found the enterprise uses a mix of approaches: on premise, 64 percent; private cloud, 50 percent;  and public cloud, 77 percent.

As for containers, most companies are deploying them to AWS, 63 percent (down from 69 percent), followed by on premise servers, 43 percent (down from 51 percent); Google Cloud Platform (35 percent down from 39 percent); Azure, 29 percent (up from 16 percent); VMware 24 percent (no earlier data for comparison); and OpenStack, 20 percent (down from 22 percent).

Although the CNCF doesn’t spell it out, these results show that many, if not most, companies are using hybrid cloud approaches.

To manage those containers, as you’d guess, Kubernetes leads the way with 83 percent, followed by Amazon Elastic Container Service (24 percent), Docker Swarm (21 percent), and Shell scripts (20 percent). The CNCF serverless event data project CloudEvents is also gaining traction, with 79 percent of the respondents to the  survey  testing it, while 21 percent are already using it in production.

Put it all together and you see cloud moving from simply replicating the data center to technologies that are finding their own new programming models.


Speak Your Mind


This site uses Akismet to reduce spam. Learn how your comment data is processed.