On-Demand Instance
A standard cloud instance model that is available immediately, has a fixed price and is not forcibly reclaimed, suited to production services requiring high stability.
Timeline (last 90 days)
Decision rules for judging whether Spot is suitable: batch inference, checkpoint-supporting or stateless workloads, tolerance of 5-10 minute interruptions, and experimental scenarios where cost comes first are suitable for Spot; latency SLAs for real users, model loading exceeding 2 minutes, US and Europe peak hours, and interruption costs exceeding the savings are suitable for on-demand instances.
On-demand instances are the standard model that is ready to use immediately, has a fixed price, and will not be forcibly reclaimed, making them suitable for production services that require high stability.
Compared to reserved instances and on-demand instances, preemptible instances occupy the lowest-price, least-stable end of the spectrum, forming the third tier of cloud computing resource pricing.
All Facts (3)
Decision rules for judging whether Spot is suitable: batch inference, checkpoint-supporting or stateless workloads, tolerance of 5-10 minute interruptions, and experimental scenarios where cost comes first are suitable for Spot; latency SLAs for real users, model loading exceeding 2 minutes, US and Europe peak hours, and interruption costs exceeding the savings are suitable for on-demand instances.
50%UnverifiedOn-demand instances are the standard model that is ready to use immediately, has a fixed price, and will not be forcibly reclaimed, making them suitable for production services that require high stability.
50%UnverifiedCompared to reserved instances and on-demand instances, preemptible instances occupy the lowest-price, least-stable end of the spectrum, forming the third tier of cloud computing resource pricing.
50%