Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 14 hours
Course Outline
Readying Machine Learning Models for Production Deployment
- Encapsulating models using Docker
- Exporting models from TensorFlow and PyTorch ecosystems
- Best practices for version control and model storage
Serving Models via Kubernetes
- Fundamentals of inference server architectures
- Implementation of TensorFlow Serving and TorchServe
- Establishing and managing model endpoints
Strategies for Inference Optimization
- Implementing efficient batching techniques
- Managing high concurrency and request loads
- Refining latency and throughput metrics
Automated Scaling for ML Workloads
- Leveraging the Horizontal Pod Autoscaler (HPA)
- Utilizing the Vertical Pod Autoscaler (VPA)
- Integrating Kubernetes Event-Driven Autoscaling (KEDA)
GPU Allocation and Resource Stewardship
- Configuration of GPU-enabled nodes
- Insights into the NVIDIA device plugin
- Defining resource requests and limits for ML tasks
Model Release and Rollout Methodologies
- Executing blue/green deployment patterns
- Designing canary rollout mechanisms
- Conducting A/B tests for model performance validation
Production Monitoring and Observability for ML
- Tracking key metrics for inference services
- Implementing robust logging and distributed tracing
- Building dashboards and configuring alerting systems
Addressing Security and System Reliability
- Hardening model endpoints against threats
- Applying network policies and granular access controls
- Safeguarding high availability standards
Recap and Future Directions
Requirements
- Proficiency in managing containerized application lifecycles
- Practical experience with Python-based machine learning pipelines
- Core knowledge of Kubernetes principles
Target Participants
- Machine Learning Engineers
- DevOps Engineers
- Platform Engineering Teams
Testimonials (1)
I like the Kubernetes part of the training as I had some tasks with K8s pods and ports and it was nice to get a better understanding now. Thank you!