- Digital Land

Performance Database Table Construction Documentation

This is documentation that explains the construction of the tables in the performance database, including details on base tables, sources of columns.

The Performance Database is designed to store and analyze performance-related metrics extracted from the source database digital_land. The tables of this DB are:

endpoint_dataset_issue_type_summary

endpoint_dataset_resource_summary

endpoint_dataset_summary

provision_summary

reporting_historic_endpoints

reporting_latest_endpoints

1. Table: endpoint_dataset_issue_type_summary

Purpose: To summarize issues associated with each dataset and its associated endpoint for various organisation.

Base Tables:
- issue: Contains records of issues related to resources sourced from the digital_land.
- resource: Holds information about the resources linked to the endpoints sourced from the digital_land.
- issue_type: Defines types of issues sourced from the digital_land.

Columns:

organisation: Extracted from the provision table.
organisation_name: Extracted from the provision table.
cohort: Extracted from the provision table.
dataset: Extracted from the provision table.
collection: Extracted from the reporting_historic_endpoints table.
pipeline: Extracted from the reporting_historic_endpoints table.
endpoint: Extracted from the reporting_historic_endpoints table.
endpoint_url: Extracted from the reporting_historic_endpoints table.
resource: Extracted from the reporting_historic_endpoints table.
resource_start_date: Extracted from the reporting_historic_endpoints table.
resource_end_date: Extracted from the reporting_historic_endpoints table.
latest_log_entry_date: Extracted from the reporting_historic_endpoints table.
count_issues: Count of issues from the issue table.
date: Current date when the query is executed.
issue_type: Type of issues from the issue_type table.
severity: Severity of issues from the issue_type table.
responsibility: Responsibility assigned to the issues from the issue_type table.
fields: Concatenated fields from issue table.

2. Table: endpoint_dataset_resource_summary

Purpose: To summarize resources associated with endpoints, including mapping and non-mapping fields. Mapping fields represent those that require conversion to align with the internal system’s accepted field names (e.g., converting ID to reference), while non-mapping fields already match the required names and can be accepted directly without modification.

Base Tables:
- provision: Contains information about datasets provisioned for a particular Organisation which include organisations, cohorts, and datasets sourced from the digital_land.
- reporting_historic_endpoints: To store historical data on endpoints sourced from the digital_land.
- column_field: This is a mapping table that connects specific columns from an endpoint to their corresponding fields sourced according to the dataset’s specification e.g UID -> reference, sourced from the digital_land.

Columns:

organisation: Extracted from the provision table.
organisation_name: Extracted from the provision table.
cohort: Extracted from the provision table.
dataset: Extracted from the provision table.
collection: Extracted from the reporting_historic_endpoints table.
pipeline: Extracted from the reporting_historic_endpoints table.
endpoint: Extracted from the reporting_historic_endpoints table.
endpoint_url: Extracted from the reporting_historic_endpoints table.
resource: Extracted from the reporting_historic_endpoints table.
resource_start_date: Extracted from the reporting_historic_endpoints table.
resource_end_date: Extracted from the reporting_historic_endpoints table.
latest_log_entry_date: Extracted from the reporting_historic_endpoints table.
mapping_field: Generated using conditional aggregation from column_field to identify fields that map correctly to columns.
non_mapping_field: Generated using conditional aggregation from column_field for fields that do not map correctly.

3. Table: endpoint_dataset_summary

Purpose: To summarize endpoint information, including the latest statuses and exceptions.

Columns:

endpoint: Extracted from the endpoint table.
endpoint_url: Extracted from the endpoint table.
organisation: Extracted from the source table.
dataset: Extracted from the resource-dataset mapping.
latest_status: Extracted from the log table.
latest_exception: Extracted from the log table.
entry_date: Entry date of the endpoint.
end_date: End date of the endpoint.
latest_resource_start_date: Start date of the most recent resource.

4. Table: provision_summary

Purpose: This table provides an overview of the provision metrics for each dataset associated with an organization, by tracking the number of active and error-prone endpoints and counting various types of issues(error, warning & notice) and endpoints.

Base Tables:
- provision: Contains information about datasets provisioned for a particular Organisation which include organisations, cohorts, and datasets sourced from the digital_land.
- organisation: Contains the names and details of organizations sourced from the digital_land.
- issue: Contains records of issues related to resources sourced from the digital_land.
- resource: Holds information about the resources linked to the endpoints sourced from the digital_land.
- issue_type: Defines types of issues sourced from the digital_land.
- column_field: This is a mapping table that connects specific columns from an endpoint to their corresponding fields sourced according to the dataset’s specification e.g UID -> reference, sourced from the digital_land.
- endpoint: Contains information about endpoints sourced from the digital_land.
- source: Provides centralized metadata for datasets across different organisations, sourced from the digital_land.
- log: Contains logs related to endpoint, sourced from the digital_land.
- reporting_historic_endpoints: To store historical data on endpoints sourced from the digital_land.

Columns:

organisation: Extracted from the provision table.
organisation_name: Extracted from the organisation table.
dataset: Extracted from the provision table.
active_endpoint_count: Represents the count of unique active endpoints for a dataset that do not have an endpoint_end_date (either NaN or blank). This calculation excludes endpoints that have been marked as retired/ended.
error_endpoint_count: Represents the count of unique errored endpoints for a dataset. This count includes endpoints where latest_status is not 200 and where endpoint_end_date is either NaN or blank, ignoring any historical errored endpoints.
count_issue_error_internal: Count of internal error issues calculated from issue and issue_type.
count_issue_error_external: Count of external error issues calculated from issue and issue_type.
count_issue_warning_internal: Count of internal warning issues calculated from issue and issue_type.
count_issue_warning_external: Count of external warning issues calculated from issue and issue_type.
count_issue_notice_internal: Count of internal notice issues calculated from issue and issue_type.
count_issue_notice_external: Count of external notice issues calculated from issue and issue_type.

5. Table: reporting_historic_endpoints

Purpose: To store historical data on endpoints, including their organization, dataset, and status.

Base Tables:
- endpoint: Contains information about endpoints sourced from the digital_land.
- source: Provides centralized metadata for datasets across different organisations, sourced from the digital_land.
- log: Contains logs related to endpoint, sourced from the digital_land.
- organisation: Contains the names and details of organizations sourced from the digital_land.
- source_pipeline: Contains information on the pipeline associated with each source sourced from the digital_land.
- resource: Holds information about the resources linked to the endpoints sourced from the digital_land.

Columns:

organisation: Extracted from source table.
name: Extracted from the organisation table.
organisation_name: Extracted from the organisation table.
dataset: Extracted from the source_pipeline table.
collection: Extracted from the source table.
pipeline: Extracted from the source_pipeline table.
endpoint: Extracted from the log table.
endpoint_url: Extracted from the endpoint table.
licence: Extracted from the source data.
latest_status: Extracted from the log table.
latest_exception: Extracted from the log table.
resource: Extracted from the log table.
latest_log_entry_date: Extracted using the max function on the entry_date from the log table.
endpoint_entry_date: Extracted from the endpoint table.
endpoint_end_date: Extracted from the endpoint table.
resource_start_date: Extracted from the resource table.
resource_end_date: Extracted from the resource table.

6. Table: reporting_latest_endpoints

Purpose: To store the most recent data on endpoints and provides the latest active endpoint data per organization and pipeline.

Columns:

organisation: Extracted from source table.
name: Extracted from the organization table.
organisation_name: Extracted from the organization table.
dataset: Extracted from the dataset table.
collection: Extracted from the source table.
pipeline: Extracted from the source_pipeline table.
endpoint: Extracted from the log table.
endpoint_url: Extracted from the endpoint table.
licence: Extracted from the source data.
latest_status: Extracted from the log table.
days_since_200: Calculated from log as days since last “200 OK” status, from subquery t2.
latest_exception: Extracted from the log table.
resource: Extracted from the log table.
latest_log_entry_date: Extracted using the max function on the entry_date from the log table.
endpoint_entry_date: Extracted from the endpoint table.
endpoint_end_date: Extracted from the endpoint table.
resource_start_date: Extracted from the resource table.
resource_end_date: Extracted from the resource table.
rn: Row number for identifying unique records.