A company needs to assign a sentiment label to 2 million archived support tickets overnight, with results needed by the next business day but no requirement for an immediate per-ticket response. Which Foundry deployment approach fits this workload?
Select an answer to reveal the explanation.
Short Explanation
This job is really just about grinding through a huge pile of tickets by the time the sun comes up, with nobody sitting there waiting on each individual answer, so the right shape for it is a batch job that chews through everything overnight as one big offline task. Building out a real-time system sized to handle all 2 million tickets arriving at once would mean paying for a ton of capacity that mostly sits idle, since nothing here is actually happening live. Reserving guaranteed low-latency capacity is solving a problem this workload does not have, because nobody needs an instant answer to any one ticket. And trying to cram every single ticket into one ongoing conversation does not match what is actually needed, which is independent labels for independent tickets, not a single long conversation.
Full Explanation
The correct answer is D. A batch inference job is designed for exactly this situation, a large volume of records that all need to be processed as one offline workload with results ready by a deadline, rather than each ticket needing an instant reply the moment it is submitted; running 2 million tickets overnight as a batch avoids paying for always-on, low-latency capacity that the workload does not need. Option A is incorrect because sizing a real-time online endpoint for 2 million simultaneous requests would be enormously over-provisioned and expensive for a task that has no live user waiting on any single response. Option B is incorrect because provisioned throughput is meant to guarantee consistent low latency for ongoing interactive traffic, a property this overnight, non-interactive workload does not require and would pay for unnecessarily. Option C is incorrect because trying to keep all 2 million tickets in a single conversational session would exceed any realistic context window and conflates a batch data-processing job with a conversational interaction, which is not what sentiment tagging of independent tickets calls for.