Fixed Schema
A design pattern in which the structure of a data table remains stable. The product side commits to keeping column names and types unchanged, with new fields appended in a backward-compatible manner, allowing customers to establish downstream dependencies in advance.
Timeline (last 90 days)
The four supporting designs that turn "writing directly into the customer's Snowflake" from a concept into an operable solution are: key-pair auth, a fixed schema, grants up front, and a freshness check.
A fixed schema means the product side commits that the structure of the audit event table will remain stable for a period, with new fields appended in a backward-compatible way rather than arbitrarily renaming or dropping columns—essentially an extension of API-contract thinking to data tables.
A fixed schema also brings version management issues: when changes are truly needed, the product side must provide a migration path, such as adding new columns while keeping the old ones, or introducing a schema version number, to avoid breaking existing queries.
All Facts (3)
The four supporting designs that turn "writing directly into the customer's Snowflake" from a concept into an operable solution are: key-pair auth, a fixed schema, grants up front, and a freshness check.
50%UnverifiedA fixed schema means the product side commits that the structure of the audit event table will remain stable for a period, with new fields appended in a backward-compatible way rather than arbitrarily renaming or dropping columns—essentially an extension of API-contract thinking to data tables.
50%UnverifiedA fixed schema also brings version management issues: when changes are truly needed, the product side must provide a migration path, such as adding new columns while keeping the old ones, or introducing a schema version number, to avoid breaking existing queries.
50%