Survey results: An update from the cloud-native community

After our insightful findings during last year’s KubeCon + CloudNativeCon North America, we were curious to get an update from the cloud-native community by talking to visitors at the European edition of the event this May. Questioning experts who came to our booth about their thoughts on Kubernetes, Prometheus and observability painted an interesting picture of the current state of cloud-native technologies and certainly provided us great learnings for our own efforts.

We are happy to share the main results of our survey with you below for a peak into the developments in the market.

Kubernetes is in production

First off, we were able to observe that many respondents are adopting more than one platform.

Managed Kubernetes is clearly on the rise with 65.2 percent of visitors of KubeCon Europe 2022 indicating to have adopted the platform – versus only 34.9 percent of KubeCon North America attendees in 2021 giving the same response. Additionally, over half of the survey participants this year are adopting native Kubernetes – another increase from just 39.5 percent last year.

Other platforms are less popular with the audience – however, especially the use of Openshift (30.4 %), OpenStack (17.4 %) and Rancher (15.2 %) has increased significantly.

While these numbers can be a testament to overall trends, they also seem to be an indication of market differences between the primarily North American respondents of last year’s survey and the mostly European participants at the recent event.

Compared to the survey responses last year, the indicated production cluster seem similar in both markets – with one exception: The number of those not using Kubernetes has been sliced in half, with only 4.3 percent not having a production cluster yet.

Otherwise, what can be noticed is that while the majority (65.2 %) has 10 or less cluster in production, the percentage of those with more than 10 cluster slightly has grown to now already 28.8 percent since our last questionnaire.

Observability: OSS vs. Ease of Use

When asking respondents which observability stack they are currently using, we were able to detect one interesting trend: Compared to last year, a lot more users seem to be on the hunt for the right solution. While during 2021’s survey only 15.1 percent were in the selection process, the number has now increased to 28.3 percent of visitors currently look for a stack.

With half of participants are still developing their own Prometheus-based solution, it does not come as a surprise that 31.8 percent simply answered that they selected their stack because it is open-source.

Just like last year, ease-of-use further remains an important criteria when choosing an observability solutions: KubeCon visitors pointed out the importance of the stack being easy to deploy, that using the solution must be simple with a quick learning curve, and that the software should be easy to use as well. Other mentioned factors when deciding which solution to implement where performance-related, concerned security, or pointed out the option to customize. Finally, the cost of the solution was the main reason for 9.1 percent of survey participants for their selection.

Implementation: Still many challenges reported

Like last year, we asked visitors again what their biggest challenges are when it comes to observability. While the order of most answered obstacles remains almost the same as last year, overall a lot more respondents reported challenges.

When it comes to their current observability stack, a total 43.5 percent face issues when it comes to scalability and the same amount finds user-friendliness to be lacking and wish for a more intuitive UI. This is up from 29.17 and 27.08 percent respectively who indicated these issues already last year.

Interestingly, 34.8 percent have a hard time due to lack of internal expertise – paired with over half of the respondents using OSS technology this certainly hints at the complexity of developing a Prometheus stack.

Other challenges have also been indicated far more often than during last year’s survey, including 28.3 percent naming the cost concerning their observability efforts as a big obstacle. The “Other” category shown in the graph above further includes challenges related to multi-cloud management and so on.

So what does this tell us about the needs of the market?

Unsurprisingly, Kubernetes remains the platform of choice and the market has now fully arrived in production. Trust in tried-and-true open-source technology seems understandably high. Many enterprises have taken on the task of developing their own OSS-based observability stack but are facing common challenges related to this complex undertaking: Scalability and overall user-friendliness of open source and Prometheus in particular remain obstacles. On the other hand, a large number of companies are lacking the internal expertise to tackle such a demanding project altogether, yet are hesitant to use a commercial provider due to the cost factor involved. All of this results in a growing number of firms actively looking for an observability solution.

This clearly indicates that the market is in need of an observability solution that builds on proven OSS yet enhances capabilities and ease-of-use while allowing users to pay only for what they need at an affordable price level that takes different budgetary concerns into consideration.

Listening to this feedback, NexClipper has developed a solution based on these market demands: We provide a full observability stack for those just starting out, yet also offer a smart observability package for companies that already developed their own Prometheus stack – basically providing observability on their existing Prometheus stack. Utilizing best-of-breed open-source technology out there, our solution includes Prometheus, Grafana, Loki, Promtail, Tempo and OpenTelementary. NexClipper Observability enhances observability with more user-friendly functionalities for alert and incident management. Additionally, our open source project Sudory can execute predefined scripts on Prometheus, making your DevOps work a breeze, while our OSS ExporterHub curates the best exporters for you (including MySQL, PromgreSQL, etc.), so teams don’t have to do the digging themselves. All of this streamlines the work of experts so that valuable time can be put where it is spent more profitably – while the per node pricing guarantees predictability and a freedom from unwanted surprises. If you would like to learn more, feel free to reach out to us!

So after all of this, we have to say a big THANK YOU to all participants in our survey and their valuable insight. They allow us to understand not only the current state of the industry but how to find a way to address their pain points effectively. It is always great to get direct feedback from users who show us which way we should set sails on our observability journey. We are happy to report, that we will continue on this course!