Course Outline
Preparing Machine Learning Models for Deployment
- Packaging models with Docker
- Exporting models from TensorFlow and PyTorch
- Versioning and storage considerations
Model Serving on Kubernetes
- Overview of inference servers
- Deploying TensorFlow Serving and TorchServe
- Setting up model endpoints
Inference Optimization Techniques
- Batching strategies
- Concurrent request handling
- Latency and throughput tuning
Autoscaling ML Workloads
- Horizontal Pod Autoscaler (HPA)
- Vertical Pod Autoscaler (VPA)
- Kubernetes Event-Driven Autoscaling (KEDA)
GPU Provisioning and Resource Management
- Configuring GPU nodes
- NVIDIA device plugin overview
- Resource requests and limits for ML workloads
Model Rollout and Release Strategies
- Blue/green deployments
- Canary rollout patterns
- A/B testing for model evaluation
Monitoring and Observability for ML in Production
- Metrics for inference workloads
- Logging and tracing practices
- Dashboards and alerting
Security and Reliability Considerations
- Securing model endpoints
- Network policies and access control
- Ensuring high availability
Summary and Next Steps
Requirements
- An understanding of containerized application workflows
- Experience with Python-based machine learning models
- Familiarity with Kubernetes fundamentals
Audience
- ML engineers
- DevOps engineers
- Platform engineering teams
Testimonials (5)
Interactivity, no reading slides all day
Emilien Bavay - IRIS SA
Course - Kubernetes Advanced
he was patience and understood that we fall behind
Albertina - REGNOLOGY ROMANIA S.R.L.
Course - Deploying Kubernetes Applications with Helm
The training was more practical
Siphokazi Biyana - Vodacom SA
Course - Kubernetes on AWS
Learning about Kubernetes.
Felix Bautista - SGS GULF LIMITED ROHQ
Course - Kubernetes on Azure (AKS)
It gave a good grounding for Docker and Kubernetes.
