The question asks for two key characteristics of virtual warehouses in Snowflake. Option B is correct because warehouses can be started and stopped at any time, providing flexibility in resource management. Option C is correct because warehouses can be resized at any time, even while running, allowing for dynamic scaling without interrupting operations. Option A is incorrect because multi-cluster warehouses must have nodes of the same size. Option D is incorrect because warehouses are billed on a per-second basis, not per-minute. Option E is incorrect because warehouses can be used for both querying and data loading operations. The community discussion with 100% consensus and upvoted comments confirms BC as the correct answer.