About Edge capabilities

An Edge capability, like Sampling or S3 synchronization, is an application that can run on an Edge site. It can access a data source to extract and process data as needed. This data can be stored in an encrypted cache to improve the security of your data and platform. An Edge capability for a specific data source runs as a job and delivers the output to Collibra Data Intelligence Platform in a secure and reliable way.

An Edge capability has a capability template that defines a specific use case, for example, data source ingestion.

Capability templates

A capability template is developed for a specific task on a specific data source type. The capability template also determines which properties are available to configure the Edge capability.

 
Capability template Description
ADLS synchronization Used to connect to Azure Data Lake Storage (ADLS)
AWS Bedrock AI

Used to integrate with Amazon Bedrock.

This capability can't be added to an Edge site that uses a MITM proxy.
AWS SageMaker AI

Used to integrate with Amazon SageMaker.

This capability can't be added to an Edge site that uses a MITM proxy.
Azure ML

Used to integrate with Microsoft Azure AI.

This capability can't be added to an Edge site that uses a MITM proxy.
Catalog Data Classification

Used to classify data from a registered JDBC data source in the Edge site.

This capability can't be added to an Edge site that uses a MITM proxy.

Catalog JDBC ingestion

Used to register a data source and synchronize schemas from a data source via a JDBC connection.

This capability can't be added to an Edge site that uses a MITM proxy.
Catalog JDBC Sampling

Used to collect and cache sample data from a data source in the Edge site via a JDBC connection.

Ensure that you meet the additional Catalog JDBC Sampling hardware requirements, in addition to the Edge site requirements.

This capability can't be added to an Edge site that uses a MITM proxy.
Collibra Protect for AWS Lake Formation

Used to set up Protect for AWS Lake Formation.

This capability can't be added to an Edge site that uses a MITM proxy.
Collibra Protect for Databricks

Used to set up Protect for Databricks. This capability appears only if the following parameter is added to the JVM configuration in Collibra Console: -Dfeature.protect.databricks=true

This capability can't be added to an Edge site that uses a MITM proxy.
Collibra Protect for Google BigQuery

Used to set up Protect for BigQuery.

This capability can't be added to an Edge site that uses a MITM proxy.
Collibra Protect for Snowflake

Used to set up Protect for Snowflake.

This capability can't be added to an Edge site that uses a MITM proxy.
DQ Connector

Used to ingest Collibra Data Quality & Observability user-defined rules, metrics, and dimensions into Collibra Data Catalog.

This capability can't be added to an Edge site that uses a MITM proxy.
GCS synchronization Used to connect to Google Cloud Storage.
Google Dataplex Catalog synchronization

Used to connect to Google Dataplex Catalog. The way to complete this capability depends on the Dataplex integration type you want to use: Dataplex ingestion or Dataplex Catalog ingestion.
Google Vertex AI

Used to integrate Google Vertex AI.

This capability can't be added to an Edge site that uses a MITM proxy.
JDBC Profiling

Used to profile and classify data from a registered data source.

This capability can't be added to an Edge site that uses a MITM proxy.
SAP AI Core Used to integrate with SAP AI Core.
S3 synchronization Used to connect to Amazon S3.
Databricks Unity Catalog synchronization Used to connect to Databricks Unity Catalog.
Technical lineage capabilities

Used to create technical lineage for different data sources. For details, go to: Add a technical lineage capability to an Edge site.

Ensure that you meet the additional Technical Lineage minimum network requirements, in addition to the Edge site requirements.

    Firewall rules so that the lineage harvester can connect to:
    • The host names of all data sources in your lineage harvesterconfiguration file.
    • All Collibra Data Lineage service instances in your geographic location:
      RegionDNS name
      aws-catechlin-ca-central-1.collibra.com
      aws-eutechlin-eu-central-1.collibra.com
      aws-sgtechlin-ap-southeast-1.collibra.com
      aws-ustechlin-us-east-1.collibra.com
      gcp-autechlin-australia-southeast1.collibra.com
      gcp-catechlin-northamerica-northeast1.collibra.com
      gcp-eutechlin-europe-west1.collibra.com
      gcp-sgtechlin-asia-southeast1.collibra.com
      gcp-uktechlin-europe-west2.collibra.com
      gcp-ustechlin-us-east1.collibra.com

      Note  We recommend that you only use DNS names in your network configurations, as the IP addresses are subject to change periodically. If you need to use IP addresses in your network configuration, we recommend using a command line utility like nslookup to query the DNS and obtain the mapping between domain name and IP address.

      Note Edge connects to different Collibra Data Lineage service instances based on your geographic location and cloud provider. If your location or cloud provider changes, Edge rescans all your data sources. You have to allow all Collibra Data Lineage service instances in your geographic location. In addition, we highly recommend that you always allow the techlin-aws-us instance as a backup, in case Edge cannot connect to other Collibra Data Lineage service instances.

You can use a man-in-the-middle (MITM) proxy between Edge and the Collibra Data Lineage service instances. For details on which data sources support the use of proxies, go to Create a technical lineage via Edge, select your data source, and see our test results in the Connect to a proxy server section.

Important While these capability templates are available for all customers, the features for which you use them might still be in beta.

Capability template structure

Each Edge capability template contains the following:

File

Description

A manifest file (YAML)

This file contains the capability metadata and input parameter requirements.

A workflow file (YAML)

This file defines the workflow and binds the parameters to capability containers.

Docker images

One or more Docker images that implement the business logic.

Note Each type of capability has its own required custom properties. These properties appear after you select a capability template from the dropdown menu.