Question

A media company runs a real-time summarization service on Amazon Bedrock using Anthropic Claude on-demand invocation in us-east-1. During a live event, request volume spiked to roughly 3x the normal peak, and the application began receiving frequent ThrottlingException errors even after implementing exponential backoff with jitter. The team confirmed their per-account requests-per-minute and tokens-per-minute quotas in us-east-1 are already at the maximum granted by AWS Support, and they cannot commit to a fixed monthly cost for Provisioned Throughput because the traffic is highly unpredictable and event-driven. They need to increase sustained on-demand capacity for these bursts as quickly as possible without a purchase commitment. Which approach best addresses this requirement?

Accepted Answer

Configure a cross-region inference profile that distributes on-demand invocations across multiple regions, aggregating the per-region quotas to increase total available throughput.

Answer

Purchase a no-commit Provisioned Throughput model unit in us-east-1 to guarantee capacity during the spikes.

Answer

Switch the workload to Bedrock batch inference so that requests are queued and processed asynchronously without hitting on-demand quotas.

Answer

Increase the retry count and lengthen the maximum backoff window so all throttled requests eventually succeed within the same region.