NVIDIA Holoscan for Media is now ready to be used in live production, taking advantage of the best of both networking and GPU technologies.
Overview
NVIDIA Holoscan for Media is a software-defined, AI-enabled platform designed for live video production, leveraging advanced networking and GPU technologies. It supports various IP media protocols and provides robust performance for high-pressure environments, making it suitable for modern media applications.
What You'll Learn
1
How to deploy applications using the Helm dashboard container
2
Why NVIDIA Holoscan for Media is suitable for high-pressure production environments
3
How to utilize NMOS Controller and Registry containers for media connection testing
Prerequisites & Requirements
- Understanding of IP media protocols like ST 2110
- Familiarity with NVIDIA DeepStream SDK(optional)
Key Questions Answered
What types of IP media protocols does NVIDIA Holoscan for Media support?
NVIDIA Holoscan for Media supports various IP media protocols, including uncompressed ST 2110, which allows it to integrate with traditional SDI setups while ensuring quality, sync, and timing across the network.
How does NVIDIA ensure the performance and reliability of Holoscan for Media?
NVIDIA guarantees performance and reliability through rigorous density and compliance assessments, incorporating ST 2022-7 redundancy and hardware failover, along with using Red Hat OpenShift 4.14 for production deployments.
What is the maximum stream capacity of the networking interface in Holoscan for Media?
Each networking interface can handle up to 35 uncompressed 1080p60 streams or eight 4K 60fps streams with a 100 Gb/s networking infrastructure, and it is also prepared for 200 Gb/s networking.
Key Statistics & Figures
Stream capacity
35 uncompressed 1080p60 streams or 8 4K 60fps streams
Each networking interface with 100 Gb/s infrastructure
Technologies & Tools
Platform
Nvidia Holoscan For Media
Enables live video pipelines and integrates AI capabilities
Container Orchestration
Red Hat Openshift 4.14
Used for production deployments to enhance stability
AI/ML Framework
Nvidia Deepstream SDK
Provides GPU-accelerated media streaming capabilities
Key Actionable Insights
1Utilize the updated deployment guide to automate network setups.The new scripts in the deployment guide allow for the automatic setup of red/blue networks and the installation of open-source services, which can significantly reduce setup time and complexity.
2Leverage the Media Gateway container for simplified media protocol handling.The Media Gateway container provides capabilities for ST 2110 ingress/egress and GPU-accelerated streaming, making it easier to develop applications without needing deep knowledge of each protocol.
3Explore the extended set of reference applications for faster development.The availability of various reference applications allows developers to quickly prototype and test their ideas, enhancing productivity and reducing time to market.
Common Pitfalls
1
Failing to account for network redundancy can lead to production failures.
Without implementing ST 2022-7 redundancy and hardware failover, systems may not withstand high-traffic conditions, risking downtime during critical operations.
Related Concepts
IP Media Protocols
AI/ML In Media Production
Nvidia Deepstream SDK