Shared Database (RDS, Aurora, Cosmos DB, etc.)

🏢 Enterprise · ☁️ MSP

1. The Challenge

A large shared database instance or cluster serves multiple business units, departments, or customers with very different query patterns and load - but the bill shows one number for the whole database, with no way to tell which unit is actually driving compute, I/O, or storage cost.

2. The Solution

Bring in per-unit database activity (query volume, IOPS, data transfer, or storage footprint) as telemetry, then allocate the database's cost by whichever measure best reflects what's actually driving spend - compute-bound workloads by query volume, I/O-bound workloads by IOPS, storage-heavy units by GB stored.

3. Dataset Needed

📘 You don't need every measure - pick the one(s) that match what's actually driving your database bill. Cost comes from your existing billing export; this dataset only supplies the usage side

4. KPIs to Build
   

   KPI	Formula	What it represents
   Cost per 1M Queries	Amortized Cost ÷ query_count × 1,000,000	Compute-driven cost efficiency per unit
   Cost per IOPS	Amortized Cost ÷ iops	I/O-driven cost share - useful for provisioned-throughput billing
   Cost per GB Stored	Amortized Cost ÷ storage_gb	Storage-specific cost per unit

5. Business Mapping to Create

   Type	Value	Purpose
   Type	Telemetry-Based Group By	-
   Rule → Condition	Condition Service Is Amazon RDS (or your DB service)	The cost slice to allocate
   Pipeline	pipline_name	Where the usage data lives
   Custom Dimension (Group By)	business_unit	The field the cost will be broken out by
   Allocation Metric	query_count (or iops, data_transfer_gb, storage_gb)	The measure that determines each unit's share - swap depending on dominant cost driver, or build separate mappings for compute vs. storage
   Time Granularity	Daily	Matches the billing period granularity
   Include Unmatched Costs	On	Costs with no matching usage go to a "Not Allocated" bucket instead of being dropped
   Mapping Name	Cost per Customer	Becomes the group-by option in Cost & Usage Explorer

6. How to Display It in the UI

   Build these as saved views in Cost & Usage Explorer, then pin them to a dashboard:

   View	Group By / Measure	What it shows
   DB Cost per Unit — Trend	Group By → Business Mapping; chart over time	Stacked chart by business unit over time
   Top Consumers by Query Volume / IOPS	Same mapping, sorted descending, table view	Flags heaviest units
   Storage Growth per Unit	Trend view isolating storage_gb	Isolates storage-cost drivers
   Drill-Down	Any unit → Next Drill by Instance/Region	Ad hoc investigation into a unit's cost spike