Workloads that frequently scale up and down within the same day—whether manually, via automation, or platform-managed—can encounter hidden cost amplification under the DTU model. When a database changes tiers (e.g., S7 → S4), Azure treats each tiered segment as a separate allocation and applies full-hour rounding independently. In some cases, both tiers may be billed for the same time period due to failover, reallocation delays, or timing mismatches during transitions.
This behavior is opaque to most users because billing granularity is daily, and Azure does not explicitly surface overlapping charges. The result is unexpected overbilling where a single database may appear to consume 28 or more “hours” of DTU in a single calendar day. While technically aligned with Azure’s billing design, this creates inefficiencies when tier switches are frequent and uncoordinated.
In the DTU-based pricing model, customers select a predefined service tier (e.g., S3, S6, S7), and are billed per hour of provisioned capacity, regardless of actual usage. If the tier is changed during the day, Azure rounds each allocation to the next full hour and may bill for both tiers if there are overlaps. Even a brief overlap or mid-hour tier switch can result in multiple billable hours for a single wall-clock hour. Over time, this behavior can result in more than 24 billed hours in a single day for a single database.
