IVS Resilience

IVS APIs use the AWS global infrastructure and is built around AWS Regions and Availability Zones. AWS Regions provide multiple Availability Zones, which are:

• Physically separated and isolated.

• Connected with low-latency, high-throughput, highly-redundant networking.

• More available, fault tolerant, and scalable than traditional single or multiple data-center infrastructures.

For more information on the APIs, see the IVS Low-Latency Streaming API Reference, IVS Real-Time Streaming API Reference, and IVS Chat API Reference. For more information on AWS Regions and Availability Zones, see AWS Global Infrastructure.

Amazon IVS Video Data Plane

Video ingestion and distribution run over Amazon IVS’s Content Delivery Network (CDN). The CDN is specialized and highly tuned for low-latency video. This enables Amazon IVS to provide customers with end-to-end, high quality video served to a global audience with minimal delay. The video CDN has global Points-of-Presence (PoPs), allowing broadcasters and viewers to be geographically dispersed.

Regardless of the AWS region where you chose to configure your Amazon IVS resources:

• Streamers automatically ingest video to a PoP geographically close to their location.

• Viewers stream video via the global video CDN.

Once ingested, video streams are processed and transcoded in one of several Amazon IVS datacenters. Amazon IVS does not provide automated failover for ingestion or transcoding failures. Instead, streamers should configure their encoders or broadcasting clients to automatically re-ingest on any broadcasting failures.