Staging Table Snapshot
Generate allows states to create a backup or “snapshot” of staging tables. This is an optional utility that can be executed as needed or embedded into the State’s ETL workflow logic.
Last updated
Generate allows states to create a backup or “snapshot” of staging tables. This is an optional utility that can be executed as needed or embedded into the State’s ETL workflow logic.
Last updated
The Generate Staging Table Snapshot Utility (“Snapshot Utility”) provides a method to create a backup copy of staging tables for future use and reference after an ETL has populated Generate’s staging tables.
To ensure consistency across all EDFacts reports for a given year, it's recommended to retain and reuse the Directory data collected for each submission year. This approach prevents the introduction of errors or inconsistencies that might arise from reloading and updating the Directory data with new organizations or changes to existing ones that were not part of the initial report.
Developers can preserve data in staging tables across ETL executions. This practice facilitates the comparison of changes and the validation of ETL logic.
The Staging Table Snapshot Utility is not available in the Generate user interface – it can only be used withing SQL Server Management Studio (SSMS).
The Snapshot Utility would be executed AFTER running an ETL to populate Staging tables.
To execute the Snapshot Utility, run the following command in SSMS with the desired parameters:
The @SchoolYear value should correspond to the data that currently exists in Staging tables.
The Snapshot Utility will make backup copies of all staging tables that pertain to the @ReportCode and place them in the Source schema in the Generate database. For example, for @ReportCode = 'C029' these tables are:
Staging.K12Organization
Staging.OrganizationAddress
Staging.OrganizationFederalFunding
Staging.OrganizationPhone
Staging.OrganizationProgramType
Staging.StateDetail
The tables are an identical copy of the Staging table(s), with the addition of three columns:
SnapshotReportCode – the report code for which the snapshot data applies
SnapshotSchoolYear – the school year for which the snapshot data applies
SnapshotDate – the date the snapshot was created
These additional columns provide features to retain backups for multiple school years, and to retain a backup for multiple report codes in the same table.
Example: After running your Child Count ETL, you could run the Snapshot Utility for C002 (a Child Count file) for 2023 to retain data from
Staging.K12Enrollmentand other related staging tables. Later in the year after running your Exiting ETL, you could run the Snapshot Utility for C009 (Exiting) to retain data from the staging tables. Both sets of data (Child Count and Exiting) will be available in each snapshot table, allowing you to review and compare if needed.
Note: The Snapshot Utility retains a single instance of data in each table for a particular School Year and Report Code. This means you cannot retain multiple “versions” of data for K12Enrollment for 2023 for C002.
For example, while you could have Snapshot data in Source.K12Enrollment for 2023 for C002, C089, C009, etc., you cannot have data for 2023 C002 “Version 1”, “Version 2”, etc.
If a Snapshot table does not yet exist for a Staging table, it will be created. If a Snapshot table already exists for a Staging table, then the data in the table will either be appended or replaced, depending on if data already exists in the Snapshot table for the specified School Year and Report Code.
The Snapshot Utility could be coded within an ETL to automatically make a backup copy of the related Staging Tables each time the ETL is executed. The following diagram shows how this feature might be leveraged at a State to reuse a Directory snapshot for Child Count reporting.
In the Source-to-Staging_Directory ETL, the Snapshot Utility could be embedded to run automatically to create a copy of all Directory data.
The Snapshot Utility will run automatically and create/update backup copies of the Staging tables in the Source schema.
In the Source-to-Staging_ChildCount ETL, required Organization information can be pulled directly from the Snapshot tables rather than from the Directory source system. This not only assures that Directory information used for Child Count will match the submitted C029 data, but may improve ETL performance by not requiring the ETL to jump to another system/database to pull Directory data.
Below is an example of code that could be used to restore data from a Snapshot table into a Staging table. In this case, data from Source.K12Organization for a school year for C029 will be restored to Staging.K12Organization for use with Child Count.
This documentation will be updated as changes are made to Generate and the utilities within Generate.
