If you are attending the upcoming KubeCon + CloudNativeCon Europe 2025 conference in London, check out the CNCF End User Technical Advisory Board (TAB) member recommendations for their top talk recommendations! 🙂
Ahmed Bebars:
With the rich and captivating topics at KubeCon + CloudNativeCon Europe 2025, it is difficult to narrow down on just a few sessions for attendance. Nonetheless, I consider these two a “must”:
Why I’m Interested:
LinkedIn must enable “bare-metal” servers to serve as a computing platform for applications; however, integration is key to flexibility. Insight from Ahmet Alp Balkan and Ronak Nathani on how challenges of tenant isolation, resilience, and developer usability are solved at such massive scales is crucial to understanding large-scale infrastructure transformation.
Why I’m Interested:
Some of the world’s largest systems contain a treasure trove of knowledge, and Artur Bergman manages them. His attention to detail with resilient and recovery-focused systems, specifically those intended to prevent cascading failures and enable rapid recovery, is important when building robust and scalable systems.
These sessions are unparalleled for those curious about how to build and manage large-scale, resilient infrastructures due to the immense focus and solutions straightforward challenges operating at scale deal with.
Alolita Sharma:
There are a lot of excellent talks at every KubeCon + CloudNativeCon conference. Some of the talks making it to my must-attend list this year include:
Town Hall Panel Discussion: End User Technical Advisory Board (TAB) Town Hall
Why? Come meet the CNCF’s end-user TAB! Learn about the TAB’s initiatives on reference architectures, project health and feedback. And use this opportunity to ask questions about areas where and how end-users can get involved.
Project Lightning Talk: Perses Update – Augustin Husson, Maintainer
Why? Perses is a project that has the potential of changing the current open source observability landscape with the development of a new and unified UI framework that will present observability analysis dashboards and results out of the box. This talk will focus on project progress, features and roadmap which is key for end-users looking to have a cloud-native choice for observability user interfaces.
Why? AI/ML continues to make deeper inroads in analyzing telemetry data and understanding systems at scale. eBay, an end-user building cloud-native observability solutions will be talking about how they have used AI algorithms to understand their telemetry metrics, traces and logs to create more predictable and accurate AI enabled observability.
Why? This end-user talk dives into the benefits of using pre-trained ML models for day one anomaly detection. Using lightweight, unsupervised pre-trained models using cloud-native tools like Kubeflow for model fine-tuning.
Why? Another interesting end-user talk takes a look at using exemplars effectively in an automated way by collecting eBPF data with OpenTelemetry and correlating application and kernel telemetry data to understand system performance.
Unifying query languages is key in reducing toil for app developers and end users to query and analyze observability data. A common query language that can leverage all observability data such as metrics, traces, profiles, events, logs to facilitate correlation, support trend analytics and provide end-to-end observability for AI applications. End users from Netflix and Apple have been driving the Observability TAG QLS workgroup to define and finalize a semantic query language spec in 2025 and are recommending SQL as a basis with further experimentation on syntaxes. This talk will explore the trade-offs between simplicity, expressiveness, and performance. This query language convergence for end-to-end analytics could enhance reliability and operational efficiency for SREs and your app developers. This could be a win-win for all end-users.
Kenta Tada:
There are many excellent sessions at KubeCon + CloudNativeCon Europe 2025, addressing key challenges and innovations in cloud-native adoption. As organizations continue to scale their cloud infrastructure, efficient observability and extensibility are becoming critical for maintaining performance and reliability. I’d like to highlight two sessions that offer practical insights for end-user organizations on leveraging eBPF—one for enhancing userspace extensibility and another for gaining deep, network visibility without infrastructure modifications.
Session: eBPF and Wasm: Unifying Userspace Extensions With Bpftime – Yusheng Zheng, eunomia-bpf
Why I’m Interested:
As end-user organizations seek to extend and customize their cloud-native environments, both eBPF and Wasm offer compelling solutions—eBPF for kernel-level optimizations and Wasm for userspace sandboxing. However, both face challenges when userspace extensions need to interact deeply with host applications. This session by Yusheng Zheng introduces bpftime, a novel framework that extends eBPF’s capabilities into userspace without the overhead of Wasm’s sandboxing. If your organization is looking for a high-performance and secure way to extend applications using uprobe and XDP in a userspace eBPF runtime, don’t miss this session!
Why I’m Interested:
For end-user organizations, gaining network visibility without disrupting existing infrastructure is a major challenge. This session by Mario Macías and Marc Tudurí from Grafana showcases how eBPF enables lightweight, non-invasive network monitoring across layers L3 to L7 without requiring infrastructure modifications. Their approach integrates low-level network telemetry with Kubernetes metadata, providing actionable insights into service communication and traffic patterns. If your organization is looking to enhance observability while minimizing overhead, this session is highly recommended!
Ricardo Rocha:
KubeCon +CloudNativeCon keeps delivering, in size and in quality. The London event will again keep a strong focus on end users, with a large number of sessions reporting experiences and best practices for a variety of use cases. These are usually my favorite sessions, together with those where community members from different vendors and organizations got together to tackle very hard problems. Hard to pick but here goes one example from each of these two categories.
Why I’m Interested:
With the hype and demand still strong, the usage of accelerators for AI and other workloads keeps growing. But looking at the longer term, the success of these new ways of computing at scale will require more efficient, sustainable use of these expensive resources. The DRA effort has shown how our community can gather and work together to deliver now the features that will help us succeed in the future. This kubecon includes several other talks about Dynamic Resource Allocation (DRA), including some from my organization.
Why I’m Interested:
Being a huge LEGO fan I’m far from being alone in our community. Hearing about the challenges faced by the LEGO group scaling out platforms and teams makes for a very interesting and entertaining session, with lessons to be learned by end users from all domains. With great speakers as well, this will be a session not to be missed.
If you’re not registered yet, there’s still plenty of time to catch these sessions – and many more – in person. Also, come say Hi and meet us at the TAB session on Friday April 4 2025. See you there!