Usage and Billing View

sys.information_schema.instance_usage is a system view in Singdata Lakehouse that records resource consumption and billing details at the instance level. Each row represents the consumption of one SKU during a single billing period (hourly or daily), making it the primary source for bill reconciliation, cost attribution, and usage trend analysis.

Data is retained from instance creation onward.

---

Field Reference

Field	Type	Description
account_id	int	Account ID
account_name	string	Account name (i.e. the instance name)
instance_id	int	Instance ID
region_name	string	Cloud region, e.g. Alibaba Cloud - East China 2 (Shanghai)
sku_category	string	SKU category — see classification table below
sku_name	string	Specific SKU name
workspace_id	string	Workspace ID
workspace_name	string	Workspace name
measurement_start	timestamp	Start of the billing period
measurement_end	timestamp	End of the billing period
measurements_unit	string	Unit of measurement, e.g. yuan/cru, yuan/GiB/day, yuan/gb, M Tokens
measurements_consumption	double	Actual consumption during the period (in the given unit)
price_rate	string	Unit price as a string, e.g. "0.020000"
amount	double	Gross amount before discount (consumption × unit price)
discount_rate	double	Discount rate: 1 means no discount, 0.8 means 20% off
total_after_discount	double	Net amount after discount (the actual billed amount)

---

SKU Categories

sku_category	sku_name examples	Description
compute	GP Virtual Cluster, AP Virtual Cluster, Integration Virtual Cluster, Bulk Ingestion, Stream Ingestion, IGS Service, Task Scheduling	Compute resource consumption, unit: yuan/cru
storage	Managed Storage, Retained Managed Storage, Job Temp Storage, Managed User Volume Storage	Storage usage, unit: yuan/GB/day or yuan/GiB/day
network	Query Internet Data Transfer	Public internet egress, unit: yuan/gb
ai	AI model calls (multiple models, input/output billed separately)	AI function consumption, unit: M Tokens

---

Query Examples

Total cost by SKU category over the last 7 days

SELECT sku_category, SUM(measurements_consumption) AS total_consumption, SUM(amount) AS amount_before_discount, SUM(total_after_discount) AS total_cost FROM sys.information_schema.instance_usage WHERE measurement_start >= CURRENT_DATE() - INTERVAL 7 DAYS GROUP BY sku_category ORDER BY total_cost DESC;

Monthly cost ranking by workspace

SELECT workspace_name, SUM(total_after_discount) AS total_cost FROM sys.information_schema.instance_usage WHERE measurement_start >= DATE_TRUNC('month', CURRENT_DATE()) GROUP BY workspace_name ORDER BY total_cost DESC;

Daily cost trend for a specific workspace

SELECT DATE(measurement_start) AS date, sku_category, SUM(total_after_discount) AS daily_cost FROM sys.information_schema.instance_usage WHERE workspace_name = '<your_workspace>' AND measurement_start >= CURRENT_DATE() - INTERVAL 30 DAYS GROUP BY DATE(measurement_start), sku_category ORDER BY date DESC, daily_cost DESC;

CRU consumption breakdown for compute clusters

SELECT workspace_name, sku_name, DATE(measurement_start) AS date, measurements_consumption AS cru_hours, total_after_discount AS cost FROM sys.information_schema.instance_usage WHERE sku_category = 'compute' AND measurement_start >= CURRENT_DATE() - INTERVAL 7 DAYS ORDER BY date DESC, cost DESC;

Workspaces exceeding a storage threshold

SELECT workspace_name, sku_name, DATE(measurement_start) AS date, measurements_consumption AS storage_gib FROM sys.information_schema.instance_usage WHERE sku_category = 'storage' AND measurements_consumption > 100 AND measurement_start >= CURRENT_DATE() - INTERVAL 7 DAYS ORDER BY storage_gib DESC;

AI function token consumption

SELECT workspace_name, sku_name, SUM(measurements_consumption) AS total_tokens_m, SUM(total_after_discount) AS total_cost FROM sys.information_schema.instance_usage WHERE sku_category = 'ai' GROUP BY workspace_name, sku_name ORDER BY total_cost DESC;

---

Notes

• storage SKUs are measured at daily granularity; all other categories are measured hourly.
• Data is not real-time — there is approximately a 4-hour delay before records appear.
• price_rate is a string type. Cast it before arithmetic: CAST(price_rate AS DOUBLE).
• total_after_discount is the final billed amount. amount is the pre-discount gross. The difference is the discount applied.