The Kubernetes Scheduler is responsible for deciding which nodes to run new pods on based on resource requirements, node availability, and various policies.

Kubernetes provides self-healing capabilities, enabling it to automatically replace or reschedule containers if they fail, ensuring minimal downtime during deployments.

AWS Lambda is a serverless compute service. It scales by adjusting the number of functions it runs in parallel, not instances, threads, or containers.

Orchestration in a hybrid cloud is facilitated by Hybrid Cloud Management Platforms (CMP), which provide tools to manage and integrate services across multiple cloud environments.

Cloud bursting is the most suitable strategy for an e-commerce platform with traffic surges. It allows the platform to scale dynamically to the cloud to handle peak loads while minimizing the migration effort.

A pre-trained model refers to a machine learning model that has already been trained on a large dataset and is available for use without further training. Typically, these models are trained on publicly available data.

A "Public Cloud" is a cloud infrastructure made available to the general public by service providers. It is shared by multiple organizations and is typically accessed over the internet. Public clouds offer scalability and cost-efficiency.

AWS Elastic Transcoder is specifically designed for video transcoding tasks. It automatically transcodes videos when new files are uploaded to an S3 bucket, making it a suitable choice for the media company's automation needs.

Container registries like Docker Hub or Google Container Registry store Docker images, which are key for deploying containers in the CI/CD pipeline.

A Private Cloud offers more control over security and compliance because it's dedicated to a single organization, allowing customization and stricter governance.