A retail company is deploying a new order-processing database on Azure SQL Database. The database is expected to grow rapidly to several terabytes within a year. Traffic is highly unpredictable, with long idle periods overnight and sudden spikes during flash-sale events. The architecture team wants automatic scaling of compute during spikes, the ability to pause compute during idle periods to reduce cost, and support for databases that can grow beyond 4 TB without a full data reload. Which deployment configuration should you recommend?