ContainerDays Hamburg 2026
as the organisers put it
Join us at ContainerDays Hamburg 2026 to immerse yourself in the latest trends and innovations in a wide range of technology-related tracks such as kubernetes, containers & DevOps but also AI, ML, GoLang Cloud Native & Platform En gineering topics.
| title | author | length |
|---|---|---|
| Stateful Workloads at Scale – Extending Kubernetes for Running OLAP Databases as a Service. | Marvin Beckers, Jayme Bird | 27m |
Runs ClickHouse for several thousand customers on Kubernetes using MultiSTS (one StatefulSet per replica) and a Make-Before-Break process that creates new… | ||
| Snapshots Aren’t Backups: The Blast Radius Problem in Kubernetes DR | Pietro Piutti, Andreas Kotes, Joep Piscaer, Florian Fröhlich | 44m |
CSI volume snapshots are fast and cheap but CSI snapshots live in the same blast radius, so they often fail as disaster recovery. The panel argues snapshot ≠… | ||
| The Airbus AI Challenge, powered by FACIS | Stéphane Poulain, Andreas Weiss | 08m |
Airbus and FACIS propose an agentic five-step maintenance flow that claims to cut $40B in ground-delay waste, eliminate $5B of misdiagnosed repairs, and enable… | ||
| Why is it so Hard to Run a 5G Core on Kubernetes—and What Needs to Change for 6G | Joel Studler, Martin Zieroth, Stefan Prodan, Anders Fagerlind, Gergely Csatari | 44m |
Telco CNFs often stay boxy: vendors ship thousands-parameter Helm charts, bundle Kafka/databases per function, and demand root/privileged containers, which… | ||
| Sovereign by Design, Open by Default | Christian Hüning, Niklas Voss | 40m |
Claims that strict sovereignty improves open source by forcing robustness and reuse; a year-old coalition grew to 11+ projects and members including SAP and… | ||
| Engineering for Productivity – DevOps Was Just the Beginning | Stephan Stapel | 35m |
DevOps established flow, feedback and learning and moved ownership into teams, but Kubernetes, microservices and toolchains raised cognitive load. Claim: Genai… | ||
| Bridging Baremetal and Virtualized Clouds: Operating Telco-Grade Kubernetes at Scale | Kashif Khan, Ganesh Vasudevan | 34m |
Ericsson describes a Kubernetes-native platform that unifies bare metal, OpenStack (via CAPO) and KubeVirt virtualization using Cluster API and Metal3.io… | ||
| One Pipeline, Three Regulator - SBOM Compliance for the Developer | Marcus Ross, Christian Nils Schneider | 35m |
Claims a single pipeline meets CRA Annex I, NIS‑2 Article 21 and ISO27001 clauses by engineering: sign commits with Sigstore gitsign (relying on Rekor… | ||
| Beyond Traditional VPNs: WireGuard Mesh Networking for Kubernetes with Tailscale | Cecil Wöbker | 14m |
WireGuard mesh replaces hub-and-spoke VPNs here by exposing Kubernetes services with a single annotation: the Tailscale Operator registers services and… | ||
| Q&A | Kelsey Hightower | 44m |
Argues Kubernetes is powerful but flawed because its declarative API only records desired state and cannot reconcile real world drift, warns against using… | ||
| PostgreSQL Major Upgrades: Swapping the Elephant Without Breaking the Room | Anton Borisov | 32m |
Built an automated blue-green upgrade framework that moves ~100 production Postgres instances from 12 to 17 with logical decoding, controlled WAL overlap and… | ||
| Taking Back Control: Building Cloud-Agnostic Container Platforms for Regulated Environments | Jinhong Brejnholt, Per Hedegaard Christiansen | 34m |
Saxo Bank builds a cloud‑agnostic Kubernetes platform using Cluster API for vendor portability, GitHub + Flux for GitOps, and custom operators plus an internal… | ||
| Building a sovereign, vendor-agnostic Cloud API | Kevin Stellmacher, Cyrill Berg | 35m |
A sovereign, vendor‑agnostic Cloud API implements a single declarative spec that deploys identical VM, networking, storage and cluster resources across stacks… | ||
| Zero Trust for APIs: From Edge to Mesh with Istio | Mofesola Babalola, Hannah Olukoye | 29m |
Shows how to enforce Zero Trust inside Kubernetes by running Istio ambient mode with its Z tunnel, CNI and optional waypoint proxy to get mTLS and JWT-based… | ||
| Everything You Need, Nothing You Don't: How Do You Actually Build a Lean Platform | Natalie Godec, Tim Urlaub, Joep Piscaer | 43m |
Panelists show how to audit and shrink platforms with concrete wins and failures: teams rip out a useless resource‑module layer from a three‑tier Terraform… | ||
| Ctrl+Alt+Deploy: Windows Workloads as a Service powered by KubeVirt | Nicolai Ort, Tobias Schneck | 38m |
Demonstrates running Windows VMs on Kubernetes using KubeVirt with an automated pipeline (KCP/KDP) that deploys images, attaches PVCs, exposes RDP via a load… | ||
| Modern Architecture in the Cloud of 2026 | Marius Zaharia | 37m |
Shows an Azure-centric set of blueprints: a WordPress deployment on AKS with ACR, Key Vault over private endpoints and Front Door for global WAF/CDN, and… | ||
| But we have GPUs at home - on-prem ai with kubernetes, Openstack & KubeVirt | Maik Zimmermann, Nicolai Ort | 31m |
Company runs on-prem GPU clusters using bare metal, OpenStack and KubeVirt with the Nvidia GPU Operator; passthrough works but wastes whole cards, vGPU and MIG… | ||
| Kube-Oddities - the quirks that keep Kubernetes interesting | Marcus Noble, Márk Sági-Kazár | 29m |
Sidecar containers are actually implemented as a special init with restartPolicy: Always, not a new field; Kubernetes resolves image references by SHA even if… | ||
| Building Agentic-Ready Infrastructure with kcp Workspaces and API Export | Abubakar Siddiq Ango | 35m |
Shows that kcp workspaces provide workspace-level API isolation and that Google ADK can auto-discover kcp OpenAPI specs so an agent sees only bound APIs. Uses… | ||
| The Evolution to Platform Engineering 2.0 | Bjoern Brundert | 31m |
Survey data shows 90% adoption of platform engineering and 76% with dedicated teams and argues for five concrete pillars: an AI-native platform that supports… | ||
| Migrating from kOps to EKS with Minimal Developer Involvement | Clifford Sanders, Adeel Ahmad Chaudhry, Florian Schreiber | 36m |
A platform team migrates eight kOps clusters to Amazon EKS by copying deployments and routing traffic through a custom EKS migration proxy built on Skipper so… | ||
| One Platform Could Not Fit Them All | Artem Lajko, Annika Opitz | 36m |
A retailer builds an Open Source Platform Framework called Cubera that packages entire Kubernetes platforms (not just apps) and wires Argo CD, External… | ||
| From Code to Clinic — Building an Internal Developer Platform for Clinical AI | Malte Groth, Moon Kim | 30m |
Institute for AI in Medicine (IKIM) turns clinical questions into a deployed platform: they run real-time ETL into FHIR (over 3 billion data points, new… | ||
| TOPF: Open-Sourcing PostFinance's Tool to Migrate and Manage Talos Linux Clusters. | Clément Nussbaumer, Sebastian Stephan | 31m |
PostFinance built TOPF (Talos Orchestrator by PostFinance), a Go CLI that migrates kubeadm clusters to Talos and manages lifecycle tasks: bootstrapping, Go… | ||
| AutoPiloting Kubernetes: how we built autonomous Control Planes | Dario Tranchitella, Gianluca Mardente | 35m |
Argues that GitOps and static desired-state hit limits at fleet scale and shows an AutoPilot model that decouples control planes from infrastructure using… | ||
| Platform Obesity, not Complexity, is killing our platforms | Joep Piscaer | 37m |
Claim: platforms fail from accumulated bloat not inherent complexity, causing cognitive overload, operational friction, and stalled ROI. Concrete evidence… | ||
| Bootable Containers: An entire OS as a Containerfile | Robert Guske | 33m |
Builds a full CentOS Stream 9 OS plus kernel from a Containerfile, uses bootable container tooling and Podman to create an OCI image, converts it to a qcow2… | ||
| Saving Money Exposed Our Architecture - Escaping ClusterAPI with Karpenter | Mathias Gebbe | 31m |
Migrates 19 clusters (≈680 nodes, ~8,000 pods) off ClusterAPI into a Terraform-managed Amazon EKS layout and replaces controller-heavy node management with… | ||
| Scaling Kube-Apiserver to Thousands of Tenants – Lessons Learned From kcp | Simon Bein | 38m |
kcp runs virtual‑user load tests simulating 10,000 workspaces with a custom framework (couldn’t use PerfTest) and sharded etcd to avoid single‑node… | ||
| Space Invaders for Kubernetes: Consolidating 50 Clusters at Scale | Juan Peña Rivera, Enrico Heine | 33m |
Company consolidates ~50 single-tenant clusters into three shared multi-tenant EKS clusters, cutting standing costs that made >60% of clusters more expensive… | ||
| The Full Stack Preview: Kubernetes, ArgoCD, and Ephemeral Databases | Cecil Wöbker | 34m |
Built per-PR full environments using ArgoCD ApplicationSet PR generator, each provisioning app pods plus a single-member mongod replica (Atlas local image)… | ||
| Optimizing Kubernetes for Low-Latency and Real-Time Containers | Antti Kervinen, Feruzjon Muyassarov, Alexander Kanevskiy | 36m |
Shows how NRI + balloons plugin in containerd and CRI‑O actually reconfigures OCI metadata, pins CPU sets, allocates CPU/memory pools, moves IRQs, and toggles… | ||
| Not All Managed Kubernetes Are Created Equal: A Deep Dive into Control Plane Architectures | Artem Lajko, Nick Berthold | 36m |
Compares control plane patterns (virtual, hosted, dedicated) and evaluates tradeoffs using Gardener, Kamaji and Cluster API; demonstrates a live Gardener demo… | ||
| Virtual Clusters in the Real World: Lessons from k3k’s Journey to Production | Rossella Sblendido | 37m |
k3k is a production-ready virtual cluster implementation that ships three modes: k3s nested control planes for strong isolation, an agentless shared mode that… | ||
| GitOps at Scale: Why Your Hub-and-Spoke Architecture is a Security Risk | Artem Lajko, Gianluca Mardente | 32m |
Centralized Hub‑and‑Spoke GitOps creates a single God-mode superuser credentials risk and runs into scaling limits: Argo CD hubs begin failing when caches hit… | ||
| Cloud Native Network Observability: Treating Routers like Kubernetes Resources | Florian Löhden | 31m |
Replaces poll‑based SNMP with gNMI streaming telemetry at 5 second samples across a five‑router spine/leaf lab and streams into Prometheus and Grafana, while a… | ||
| From “No Time For GitOps” To Enterprise Adoption: Selling Flux The Human Way | Lucas Hornung, Christian Matthaei | 27m |
At Tchibo, platform work on Kubernetes (GCP) with Terraform and Flux moves from ignored to mandated after reframing GitOps as measurable business value: a… | ||
| From Frankenstein to Kamaji: Lessons in Building a Single CAPI Cluster Across Multiple Providers | Xavier Avrillier, Antonia von den Driesch | 37m |
Giant Swarm prototypes a hybrid Cluster API cluster with the control plane on vSphere and workers on Proxmox, which initially requires a Frankenstein cluster… | ||
| Fun with Flags: How OpenFeature Solves Your Feature Flag Headaches | Simon Schrottner | 38m |
A vendor-agnostic API called OpenFeature provides language SDKs and four concrete tools: the evaluation API, pluggable providers, an evaluation context for… | ||
| Rooted in Open Source: How CNCF Projects Mature and Scale | Jeffrey Sica | 40m |
Describes how projects move from Sandbox to Graduation using a real example, Alpaca, with concrete requirements: Apache 2 license, transfer of trademark and… | ||
| Running PostgreSQL on Kubernetes in Production: Day 2 Operations with CloudNativePG | Michael Raeck | 30m |
CloudNativePG operator reliably provisions a primary plus read replicas and auto-creates secrets, but continuous backup not configured by default so Barman… | ||
| Multi-tenant K8s Networking with Kube-OVN | Tom Schwaller | 33m |
Kube-OVN plus Kubermatic Virtualization implements VLAN- and overlay-based VPCs on a layer‑2 ISIS fabric across three AZs, wiring provider bridges to bond… | ||
| Building Production-Ready Kubernetes Operators | Fortune Ndlovu | 39m |
Red Hat Developer Hub Operator manages Backstage using operator-sdk and controller-runtime in Go, merging a default config with the CR spec, applying plugin… | ||
| Kubernetes RBAC at scale without losing your sanity | Kevin Gimbel | 30m |
AOE uses Kubernetes’ built‑in aggregationRule on ClusterRole to deploy many tiny cluster roles labeled for specific CRDs or verbs and have them automatically… | ||
| Rick Rackow | Rick Rackow | 33m |
A small Go program moves from serial to concurrent to parallel code and the demos show concrete timings: the binary runs ~2.9s on the host and inside Docker… | ||
| Specs Don't Hallucinate: Deterministic Code in the Age of AI | Artur Iablokov | 29m |
An experiment builds a Go task service using oapi-codegen and sqlc versus pure AI generation across greenfield and brownfield scenarios. Ten greenfield runs… | ||
| But can you really run your app on 2 clouds at the same time? | Natalie Godec | 40m |
A production platform serving images runs simultaneously on AWS and GCP with zero downtime by deploying Kubernetes clusters, using Postgres with pglogical for… | ||
| From Weeks to Minutes: Automated Infrastructure Discovery for Cloud Migrations | Simon Ostendorf | 35m |
A script-based framework runs read-only shell and PowerShell collectors (plus Visual Basic for Windows Server 2008) against a host CSV, normalizes outputs into… | ||
| The Cluster Is Not the Platform: Operating Industrial Edge Kubernetes as a Fleet | Constantin Korte | 24m |
Claim: industrial edge Kubernetes becomes operable at scale by encoding per-site site profiles and exceptions, pushing them with GitOps (Argo CD) and local… | ||
| A Swiss Cloud-Native Farm 🐄 | Clément Nussbaumer | 36m |
A working on‑prem farm stack: a 4‑node bare‑metal cluster (~$350/node, 16GB each) runs Kubernetes on Talos Linux, managed with OpenTofu and GitOps (FluxCD)… | ||
| Operating a multi-tenancy Kubernetes service on KubeVirt | Fabio Berchtold | 42m |
Swisscom builds an on‑prem, multi‑tenant Kubernetes service using KubeVirt, Kube‑OVN, Trident‑CSI and Kubermatic KKP, deployed across four datacenters with one… | ||
| Kubernetes as the Universal Control Plane: From Pods to Multi-Cloud Infrastructure | Rabieh Fashwall | 36m |
Concrete claim: Kubernetes acts as a universal control plane when extended with Crossplane and Cluster API so teams use one API and kubectl/YAML to provision… | ||
| Scaling Kubernetes in Regulated Environments Through Platform APIs | Mahesh Narayan | 23m |
Siemens scales Kubernetes across 30 platform teams and decades‑old medical devices by replacing raw cluster primitives with intent‑based platform APIs that… | ||
| Tying a metrics to every release | Asaph Kotzin | 06m |
A company ties a metric to every release and shows concrete results: releases increase 68% more releases over six months while team size stays flat, every… | ||
| One update, a thousand locations.Container Management In Edge Computing Made Easy | Sebastian Strollo | 10m |
Avassa is a platform that manages containers across hundreds or thousands of retail, factory, or vehicle sites using a central control tower and a tiny on-site… | ||
| VM: When Compute Moves but Data Stays | Chris Engelbert | 06m |
KubeVirt performs live migration moves CPU and memory but not local disk by copying snapshots and deltas, freezing source, applying deltas, and resuming the… | ||
| The Reality of Rootless Containers | Ajitem Sahasrabuddhe | 35m |
Rootless containers use user namespaces with a 100000+ UID range (typically 65,536 slots) so UID0 inside maps to an unprivileged host UID, and root is a set of… | ||
| Why Isn't the Fix in My Container? Tracking CVE Propagation Across 10,000 Projects. | Lior Kaplan, Yuval Saggie | 35m |
Analysis of 10,000 open-source projects (daily scans for a year) tracks ~40,000 CVE paths and finds that 89% fixes exist but distribution lags: 40% take >180… | ||
| So I generated over 3000 SBOMS for the CNCF ecosystem | Mario Fahlandt | 36m |
A CNCF-wide pipeline generates 12,500 SBOMs from landscape-sourced releases, using Wayel for multi‑language SBOMs, GitHub Actions in 2,500+ parallel jobs, S3… | ||
| CloudNative Coffee: Building Resilient Prometheus Exporters for Flaky Edge Devices | Ronald Philipsen | 14m |
A Prometheus exporter written in Go scrapes a local gaguino API and also consumes the espresso machine’s websocket HP stream, caching state so scrapes return… | ||
| Surviving the Vulnpocalypse: Open Source and Supply Chain Security in a Post Mythos World. | Adrian Mouat | 36m |
Claim: frontier models like Mythos autonomously find and weaponize bugs, agents even hacked Hugging Face and municipal ICS, and exploitation now often precedes… | ||
| Containers Don't Keep Secrets: Scanning Docker Hub for Leaked Credentials and Private Keys. | Sebastian Strollo | 32m |
Scan of Docker Hub images processes 200 TB scanned into a BA2 normalized archive, deduplicates layers, and runs format-specific rules to yield ~100,000… | ||
| How Clusters Fail Under Attack and How We Detect It at Runtime with eBPF signals | Mohamed Ali Mellah | 34m |
Shows a practical runtime stack: eBPF programs (tracepoints, kprobes, LSM hooks) run in-kernel with low overhead (<1%) and map syscalls back to pod and… | ||
| Beyond Reconcile Loop - Patterns for Kubernetes Operator Development | David Steiman | 35m |
Three concrete tricks: use Tilt for near–production live-reload during operator development; keep YAML templates compiled into the binary via go:embed (or… | ||
| How TLS Can Get You Hacked — When Encryption Becomes Trust | Ali Alp | 33m |
Service mesh Istio, Envoy, Kubernetes service accounts issues cause identity abuse, not crypto failures: mTLS proves who a caller says it is, but authorization… | ||
| Blind Trust is Broken: Securing Your Cluster Against Untrusted Containers | Dominik Mathern | 31m |
Claims that 85% of images have high or critical vulnerabilities and recommends stopping pulls from public registries by enforcing a single source of truth and… | ||
| CI/CD is now AI/CD | Jonathan Vermeulen | 22m |
AI multiplies code output so CD becomes the bottleneck; the concrete proposal routes each change by measured risk using CD policy ops that check provenance… | ||
| Kafka Without Disks: Making Streaming Brokers Disposable on K8S | Anton Borisov | 36m |
A migration from MSK to AutoMQ moves durability into object storage durability and turns brokers into stateless brokers (no replication, no PVs), tested on… | ||
| Federated Identity for Distributed Systems: Understanding SPIFFE | Leon Krass | 36m |
Claims that SPIFFE replaces long‑lived secrets with short‑lived cryptographic identities, issues X.509 and JWT credentials via a workload API, and uses node… | ||
| NeoNephos’ OpenControlPlane: Building Europe’s Next-Gen Control Plane-as-a-Service. | Maximilian Techritz, Johannes Ott | 29m |
Small engineering team (45 full-time) builds OpenControlPlanes donated to the NeoNephos Foundation and funded by the EU, claiming 10x faster cloud landscape… | ||
| Istio's Ambient Mesh: The Real Cost of Sidecar-less Tracing | Mofesola Babalola, Hannah Olukoye | 34m |
Benchmarks compare sidecar mode to Ambient Mesh using ztunnel (Rust) + eBPF and waypoint proxies, showing lower steady CPU/memory for ambient L4 and slightly… | ||
| Beyond Tool Sprawl: An Event-Driven Approach to Dynamic Multi-Cloud Operations | Eleni Grosdouli | 24m |
Claims an event-driven on-demand service provisioning model that uses GitOps plus Sveltos to spin up Cloud Native PG instances, return credentials, and manage… | ||
| AI-Driven Incident Response: From Alert to Postmortem | Philipp Westphalen | 28m |
A Postgres migration that rewrites a large product table causes a table lock and three outages after the same version is deployed three times; rollback… | ||
| YOLO Developer Workflows with a Coding Agent in a Box | Manuel De La Peña | 34m |
Docker builds a lightweight microVM sandboxes (sbx) that run coding agents root‑inside but isolate host disk, network egress and secrets; demo shows sbx can… | ||
| From Minutes to Seconds: How we reduced ML models cold start time by 90% on Kubernetes | Simone Sciarrati, Federico Hernandez | 35m |
Claim: cold-start time drops ~90% from ~11 minutes to ≈1 minute by combining Kubernetes + KServe with node prep and image/model preloading. Changes: increase… | ||
| From Von Neumann to Transformers: An Infra Engineer's Guide to What's Actually Inside an LLM. | Stefan Schimanski, Sergiusz Urbaniak | 38m |
A 110M-parameter GPT-2 clone is trained from scratch on ~1GB of German text using a 32k-token vocabulary and 768-dim embeddings, producing German-like output… | ||
| AI Training & Inference on Kubernetes: Compute Patterns and Best Practices with Karpenter. | Christian Meléndez, Oussama KANDAKJI | 40m |
Provides concrete patterns and knobs for running training and inference on Kubernetes using Karpenter: gang scheduling and placement-group use for multi-node… | ||
| Agentic AI Under Attack: Live Demos of Exploits Through Autonomy and Trust | Nico Meisenzahl | 36m |
Four live demos show concrete attacks and mitigations: a CVE triage agent accepts an injected markdown comment and marks a critical bug CVE triage false… | ||
| The "Agentic Organisation": Governing Multi-Agent Developer Teams with Nutanix Agent Gateway and MCP | Andreas Mensing | 18m |
Central control plane enforces granular tool filtering and least-privilege access for multi-agent workflows using Nutanix Agent Gateway and Model Context… | ||
| Secure Your Coding Agent Like It’s Malware | Richard Gross | 34m |
Coding agents get treated like untrusted malware: the recommended architecture clamps them into tightly hardened containers with strict firewalls, minimal… | ||
| From Prototype to Production: Making Agentic AI Systems Reliable | Donald Forbes | 33m |
Argues that Temporal makes agentic AI production-ready by turning LLM + tool loops into durable workflows with durable execution, persisted state in a backing… | ||
| The Missing Layer in the AI Stack: Cognitive Performance for the AI Era | Hossein Rafieekhah | 32m |
Claims that human decision speed is the bottleneck and presents a deployed Cognitive Performance Layer called Neura Space that senses behavior, maps a… | ||
| Own Goal: How AI Agents Are Quietly Scoring for the Other Side | Mohamad Sanan | 28m |
JFrog data and incident analysis show real risks: eight packages leaked ~83,000 secrets, 25 packages caused ~2.5 million compromised downloads, npm malicious… | ||
| AI took over dev. Prod is next. Here's why it has to be | Dominik Rehbock | 11m |
Anthropic data shows engineers merging 8x more code since 2025; prod lags because mistakes cost uptime, error budget, and regulators. An OpenAI eval agent… | ||
| The Human is the Next Bottleneck in the AI Stack! | Hossein Rafieekhah | 07m |
Claims a new Cognitive Performance Layer that senses behavior, builds a digital twin, adapts AI output to current capacity and actuates tasks, estimating 40%… | ||
| Own your Agentic AI | Akshay Jain | 07m |
Claims a platform that keeps data on-prem or in a chosen cloud and runs agents on private GPUs while remaining OpenAI-API compatible. Demonstrates full audit… | ||
| From Chatbots to Agents: Orchestrating Agentic AI on Existing Enterprise Kubernetes Infrastructure. | Rene Schramowski, Ruben Verma | 27m |
Shows a working loan-chatbot stack deployed on existing Kubernetes using OpenShift AI and operators to manage model serving, Jupyter workbenches and GPUs… | ||
| Making the case for a vendor-neutral control plane to scale AI in enterprise settings. | Ali Raza, Razvan Radulescu | 22m |
Argues for a vendor-neutral control plane that centralizes skills, tools and context, demonstrates an open-source prototype Hexis wired to Claude, Granola… | ||
| MCP Security: Keep Your AI Agents from Spilling the Tea | Manfred Bjørlin, Awar Abdulkarim | 29m |
MCP security discussion shows a concrete self‑hosted setup: KAgents on AKS running version 0.9, pulling models (including Olama) and exposing ~12 MCP tools per… | ||
| Gateway API: The Illustrated Guide to the Future of Kubernetes Networking | Aleksandr Rybolovlev Alen | 35m |
Gateway API claims to replace messy Ingress annotations (some controllers had over 100 annotations) by enforcing an explicit GatewayClass–Gateway–Route… | ||
| Autoscaling Isn't Set-and-Forget: Tuning KEDA & Karpenter for Real Workloads. | Christian Meléndez, Jan Wozniak | 36m |
KEDA + Karpenter tuning focuses on concrete fixes: choose proactive metrics not CPU/memory, use KEDA scaling modifiers and the double-question-mark fallback to… | ||
| VPCs for Everyone: Multi-Tenancy and Advanced Networking with kube-OVN | Jan Bruder | 36m |
kube-OVN brings Open vSwitch + OVN into Kubernetes to provide VPC-like segmentation, subnet CRDs, and QoS for multi-tenant clusters, trading simplicity for… | ||
| Beyond “Supports OpenTelemetry”: A Maturity Model for Cloud-Native Observability | Kasper Borg Nissen | 35m |
Introduces a community OpenTelemetry support maturity model and applies it to five Kubernetes ingress controllers (kgateway, contour, emissary, traefik, istio… | ||
| CSI WON’T SAVE YOU - Why “CSI Support” is not a storage strategy and you are being setup to fail. | Daniel Paul | 32m |
CSI handles only create attach detach remove and ignores replication disaster recovery and data mobility, so relying on it leads to scale and operational gaps… | ||
| Can You Move Production vSphere VMs To KubeVirt? Yes. Here's How | Ronny Issac | 33m |
A real migration pipeline that moves a 4 vCPU, 2 GB RAM Debian VM from vSphere to KubeVirt using Forklift, streaming disk conversion with virt-v2v and creating… | ||
| Unleashing the Tides of Kubernetes Networking by Removing kube-proxy | Tomasz Tarczyński, Filip Wardzichowski | 31m |
Replacing kube-proxy with Cilium’s eBPF data path claims to restore near–bare‑metal throughput: a 100 Gbps baseline drops to ~63 Gbps with default Cilium… | ||
| Network Observability with Calico | Dillon Barry | 20m |
Calico Whisker is an open-source flow observability tool bundled with Calico (used to secure over 8 million nodes and manage 1M+ workloads) that streams… | ||
| From veth Pairs to Services: The Secret Life of Container Packets | Ajitem Sahasrabuddhe | 26m |
Follows a real packet from a pod through a Linux network namespace, creating a namespace via unshare and wiring a veth pair with Go and the netlink library… | ||
| Pods Under Pressure: Understadning PSI for K8s | Anton Weiss | 29m |
Linux PSI (available since 2018) reports sum/full/total pressure for CPU, memory and IO per process averaged over 10s/1m/5m and appears in Kubernetes 1.34… | ||
| It’s Always DNS: So Let’s Break It On Purpose | Joel Studler, Georgios Daskalopoulos | 34m |
A cloud‑native DNS for a 5G core is built with PowerDNS + ExternalDNS and CoreDNS forwarders across Kubernetes clusters, using GitOps for automation and no… | ||
| Running the Runners: How CNCF Powers GitHub Actions at Scale | Koray Oksay, Jeffery Sica | 36m |
CNCF builds a shared, self‑hosted GitHub Actions platform that runs ephemeral VMs per job on Oracle Cloud using a Kubernetes control plane with… | ||
| Ingress-Nginx: The Day After | Nicolas Mengin | 35m |
Ingress-NGINX is end-of-life and Traefik claims a practical migration path: an audit binary scans clusters and reports which of your Ingress-NGINX annotations… | ||
| Don’t Debug, Reset: Managing Kubernetes the Declarative Way with Talos Linux | Daniel Bodky | 34m |
Talos Linux is a minimal, immutable distro that runs from a squashfs image and exposes an mTLS API so node and cluster lifecycle happens declaratively via a… | ||
| Kubernetes on the edge: running low-latency workloads with WebAssembly | Rene Schach | 33m |
Demo builds a Go web server compiled with TinyGo and fetched/pushed with WKG to a local OCI registry, then runs it on a kind cluster wired by SpinCube (spin… | ||
| Networking Failure Stories: How to Maintain, Break, and Fix Cluster Connectivity | Tomasz Tarczyński | 31m |
Cilium users hit two concrete failure modes: a misconfigured network policy plus identity-label churn exhausts eBPF policy maps (cluster set to 256 entries… | ||
| D01 R01 02 Tim Urlaub | Tim Urlaub | 35m |
A platform engineering maturity model is derived from the CNCF model (50 contributors) into an internal Aldi version that adds clear target properties, an… | ||
| nothing matches | ||