roclive
Ask the service rate limite and etc.
Can the service runpod.io meet such needs:I would like to convey our usage scenario. Specifically, we are looking to provide a public network service, with initial users estimated to be around 2,000 to 10,000 (about 2,000 to 10,000 teachers from 30,000 middle schools). If each user has about 10 uses per day, that would result in approximately 20,000 to 100,000 requests. In this case, is there a possibility that runpod.io's rate limiting or circuit breaker would be triggered? Is it possible to connect directly from an Azure VM to a runpod.io GPU instance using a RESTful API without using any SSH tunnel? And we will use Ollama deployed in the runpod gpu instance
2 replies