Pepperdata has released The State of Kubernetes 2023, the results of a survey conducted with 800 C-level execs and senior ITOps and DevOps professionals in industries including financial services, healthcare, technology, and advertising. The survey aimed to gain insights into the types and sizes of containerized applications and other workloads on Kubernetes clusters, as well as the FinOps tools and practices used to optimize Kubernetes deployments and control costs.
Key findings from the survey show:
1. The Kubernetes market is maturing.
Evidence of this includes:
■ The number of clusters that are being deployed has grown to six to ten clusters per organization.
■ The variety and types of workloads that are now being launched by Kubernetes extend beyond microservices to data ingestion, cleansing, and analytics, databases, and artificial intelligence (AI) and machine learning (ML).
■ An emerging challenge has been unexpected infrastructure spend and the growing focus on cost control and optimization.
2. Companies deploying Kubernetes clusters are doing so to stay competitive or ahead of the competition, and need the agility Kubernetes affords them to deploy services rapidly.
3. Cost savings was identified as the top ROI metric for a Kubernetes deployment.
4. Companies are turning to a wide variety of approaches to attempt to remediate infrastructure spend overruns.
"The survey confirms that Kubernetes has become the preferred choice for deploying workloads among agile enterprises. However, the speed and ease of deployment can result in unexpected infrastructure cost overruns. Respondents are increasingly turning to FinOps and cloud cost optimization tools to help them remediate the cost of operating in the cloud and optimize their Kubernetes clusters," said Ash Munshi, CEO, Pepperdata. "Regardless, the survey shows that cost savings is the top metric people are using to measure the ROI of their Kubernetes investments in 2023."
Other findings included:
■ How many? The average number of Kubernetes clusters among those surveyed ranged from three to ten. 32% reported having three to five deployments, while 38% reported six to ten clusters. Almost 15% said they had between 11 and 25 clusters, and four% reported more than 25.
■ What for? When asked what types of applications were being deployed, respondents chose from a list of workloads, with the most popular being data ingestion, cleansing, and analytics, including Apache Spark as the choice for 61% of those polled. Other popular deployments included databases or NoSQL databases such as PostgreSQL, MongoDB, and Redis; and web servers like NGINX.
■ What challenges? The benefits of Kubernetes are many, but companies also find that getting a Kubernetes project off the ground comes with many challenges. These include significant or unexpected spend on compute, storage, networking infrastructure, and/or cloud-based IaaS; a steep learning curve; and a lack of visibility into Kubernetes spend, leading to cost overruns. Over 57% cited the "significant or unexpected spend on compute, storage, networking infrastructure, and/or cloud-based IaaS" as their biggest challenge.
■ Big upsides. There are many ways to measure the return on Kubernetes investments. Those polled found that cost savings was the primary metric they used to measure success. Almost 44% of the organizations are implementing cloud cost reduction and FinOps initiatives to reduce cost overruns.
The Rise of FinOps for Kubernetes Workloads
FinOps is a budding approach to managing and optimizing cloud spending across different teams within an organization. Given the increased interest in managing Kubernetes costs in the cloud, those surveyed were asked about their experience with FinOps.
The FinOps Foundation defines levels of familiarity with FinOps, from crawl to walk to run. Similar to the results identified in The State of FinOps 2022 survey, most respondents self-identified at the walk stage, meaning they can implement cloud cost reduction recommendations. The second largest group self-identified as the crawl stage, with the ability to visualize cloud costs even if they haven't yet begun the process of implementing recommendations for remediation. Interestingly, more than 98% of respondents indicated familiarity with FinOps and saw themselves somewhere on the continuum of implementing best practices for cloud cost remediation. In addition, more than 17% of respondents identified themselves in the run stage, with the ability to remediate cloud costs autonomously.
