Question

A media company runs a nightly job that summarizes roughly 500,000 news articles between 1:00 AM and 5:00 AM using Amazon Bedrock. During the day, the same application handles sporadic, low-volume interactive summarization requests from editors. The team wants to minimize total Bedrock cost while ensuring the nightly job reliably completes within its window. Which approach best meets these requirements?

Accepted Answer

Submit the nightly articles as a Bedrock batch inference job and serve the daytime editor requests with on-demand (InvokeModel) throughput.

Answer

Purchase a 1-month provisioned throughput commitment sized for the nightly peak and route both the nightly job and daytime editor requests through it.

Answer

Run both the nightly job and daytime requests entirely on on-demand throughput, relying on exponential backoff to handle any throttling.

Answer

Purchase provisioned throughput for the daytime editor traffic and submit the nightly job as a batch inference job.