Explore advanced Kubernetes scheduling techniques in this 34-minute conference talk from KubeCon + CloudNativeCon North America 2022. Discover how to optimize workload placement beyond basic CPU and memory considerations for 5G, AI/ML, HPC, and NFV applications. Learn about Kubernetes scheduling extensions that enable creative solutions for complex resource allocation. Dive into open-source projects like Telemetry Aware Scheduling and GPU Aware Scheduling, which allow for intelligent decision-making based on various metrics. Gain insights on deploying and configuring your system to handle diverse use cases, considering factors such as node busyness, GPU availability, minimal throughput requirements, and even temperature constraints. Enhance your cluster management skills and make smarter scheduling decisions for your workloads.
Working Your Cluster: Smarter Scheduling Decisions for Workloads
CNCF [Cloud Native Computing Foundation] via YouTube
Overview
Syllabus
Working your Cluster: Smarter Scheduling Decisions for Your Work... Madalina Lazar & Denisio Togashi
Taught by
CNCF [Cloud Native Computing Foundation]
Related Courses
-
Getting Started with HashiCorp Nomad
-
Precision Matters: Scheduling GPU Workloads on Kubernetes
-
Leverage Topology Modeling and Topology-Aware Scheduling to Accelerate LLM Training
-
Improving GPU Utilization and Accelerating Model Training with Kubernetes Scheduling Framework and NRI
-
Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads
-
Efficient Multi-Cluster GPU Workload Management with Karmada and Volcano
