CNCF [Cloud Native Computing Foundation] Courses

NanoVisor: Revolutionizing FaaS Cold Start Performance with Secure, Lightweight Container Runtime

Explore NanoVisor's innovative approach to enhancing FaaS cold start performance through lightweight container runtime, achieving secure sandboxed containers in just 5ms with minimal memory overhead.

• YouTube
• 26 minutes
• Self-Paced
• Free Video

Ensuring Success with Kyverno in Production - What You Must Know

Explore essential strategies for successful Kyverno deployment in production environments, covering policy creation, enforcement, scalability, and integration with existing workflows. Gain insights on optimizing performance and maintaining stability.

• YouTube
• 33 minutes
• Self-Paced
• Free Video

Developing a Standard Multi-Cluster Inventory API

Explore the standardization of cluster inventory APIs, their progress, use cases, and future vision. Learn about multi-cluster management and AI workload scheduling challenges.

• YouTube
• 35 minutes
• Self-Paced
• Free Video

Unlocking LLM Performance with eBPF - Optimizing Training and Inference Pipelines

通过eBPF实现LLM训练和推理过程的可观察性,优化性能。介绍内存、网络和GPU分析方法,分享在PyTorch和llm.c项目中的实践经验,提高GPU利用率。

• YouTube
• 38 minutes
• Self-Paced
• Free Video

How to Increase the Throughput of Kubernetes Scheduler by Tens of Times

Explore advanced techniques to significantly boost Kubernetes scheduler performance, including algorithm optimization, parallel computation, and batch scheduling for improved cluster scalability and resource utilization.

• YouTube
• 41 minutes
• Self-Paced
• Free Video

Constructing the 10x Efficiency of Cloud-Native AI Infrastructure

Explore strategies to maximize GPU utilization, unify accelerators, manage resources, and optimize AI infrastructure for 10x efficiency in cloud-native environments.

• YouTube
• 35 minutes
• Self-Paced
• Free Video

Karmada and Istio - Improving Workload and Traffic Resilience in Production Distributed Cloud

Explore how Karmada and Istio enhance resilience in distributed cloud environments, focusing on workload management and traffic distribution across multiple clusters.

• YouTube
• 37 minutes
• Self-Paced
• Free Video

The Experience of ChillyRoom Developing and Managing Session-Based Games on Kubernetes with OpenKruiseGame

Explore ChillyRoom's experience using OpenKruiseGame for session-based game development on Kubernetes, covering automated networking, elastic scaling, and room management for game developers.

• YouTube
• 43 minutes
• Self-Paced
• Free Video

Breaking Boundaries - TACC as a Unified Cloud-Native Infrastructure for AI and HPC

Explore TACC, an innovative AI infrastructure solution bridging K8S and Slurm advantages. Learn about seamless UI, multi-tenant resource management, and robust distributed infrastructure for large-scale GPU clusters.

• YouTube
• 32 minutes
• Self-Paced
• Free Video

JuiceFS CSI in Multi-Thousand Node Kubernetes Clusters for LLM Pre-Training

Explore JuiceFS CSI's innovative solutions for managing large-scale data in Kubernetes, addressing storage challenges in LLM pre-training with mount pod and sidecar modes for optimal performance.

• YouTube
• 26 minutes
• Self-Paced
• Free Video

WebAssembly on the Server - Exploring Server-Side Applications and Integration

探讨WebAssembly在服务器端的应用,包括与容器工具的集成、用例分析,以及在LLM等领域的潜力,展示其作为安全高效的云原生技术的优势。

• YouTube
• 44 minutes
• Self-Paced
• Free Video

Expanding Cloud Native Capabilities with WASM: A Case Study of Harbor

Explore WASM integration in Harbor for flexible cloud-native artifact registry extension, enabling custom filtering and security auditing without code modification.

• YouTube
• 39 minutes
• Self-Paced
• Free Video

Next Steps for the Ingress-NGINX Project

Explore the future of Ingress-NGINX, the widely-used ingress controller. Learn about recent updates and upcoming plans in light of Gateway API GA, shaping the project's evolution in cloud-native environments.

• YouTube
• 40 minutes
• Self-Paced
• Free Video

Accelerating Serverless AI Large Model Inference with Functionalized Scheduling and RDMA

Optimizing serverless AI inference with functionalized scheduling and RDMA for faster response times and efficient KV cache migration, addressing challenges in large-scale model deployment.

• YouTube
• 35 minutes
• Self-Paced
• Free Video

Beyond Statefulset: Containerizing Enterprise Stateful Applications in Practice

Explore advanced techniques for containerizing enterprise stateful applications, including leader election, hot loading, and database maintenance, with insights from Alibaba and Xiaohongshu teams.

• YouTube
• 43 minutes
• Self-Paced
• Free Video