Roadmap¶
Egeria is a large project with many activities adding content to the project. This page provides an overview of the aims of the project and a reflection of where we are today.
Capability layers¶
Egeria aims to deliver against 7 capability layers:
Governance solutions¶
The governance solutions aim to support an organization's governance efforts by providing pre-built capability not typically found in commercial tools.
The implementation of a governance solution is focused mainly on the extension of the Egeria User Interfaces and associated reference data to support additional roles and functions. They make use of the services provided by the developer platform and may exploit additional content, utilities and connector implementations from the integration platform.
Education¶
Provides educational resources for different personas and starting points.
Egeria's education aims to broaden the knowledge of people who need to work with digital resources about metadata, governance practices and the use of Egeria. It is based around the Coco Pharmaceuticals scenario and are organized by persona, so you can target your learning to your interests.
- The Egeria workbooks are part of the Egeria Workspaces deployment environments. They provide guidance and examples on how to use Egeria and how to adapt it to support your use cases. This is an excellent starting point to get hands-on experience with Egeria.
- The open-metadata-samples module in the main git repository for the Egeria code - egeria.git - includes many coding samples showing how to use the Egeria interfaces. There are also many sample clients, server configurations and sample data included in the open-metadata-deployment module. They are accompanied by
README.md
files to explain how to use them. - The Open Metadata Labs build out an operational open metadata ecosystem and show how different governance use cases can be implemented.
- This website (https://egeria-project.org/) contains comprehensive documentation on Egeria's features.
- The Governance practices provides governance best practices. They aim to guide a team that is setting up or revising their governance program through common governance tasks.
User Interfaces¶
Most users will experience the open metadata ecosystem via their own tools. However Egeria does have some simple user interfaces to cover its unique capabilities.
- Hey Egeria provide command line controlled commands and visualizations for a running Egeria system. They are designed for technical and data professional working in the open metadata ecosystem and operating Egeria infrastructure.
- Scripting commands allow calls to Egeria's services from scripting languages to aid automation of governance.
- Brain Explorers are cloud based websites allowing you to interact with a graph interface to visualized open metadata content.
- Mermaid graph visualization for visualizing collections of related open metadata elements.
Integration platform¶
Supports the integration of popular technologies into the open metadata ecosystem with pre-built connectors that are installed in the standard runtime and automatically configured and started when the appropriate content pack is loaded. Minimal coding (using the developer platform) is still required around unusual and home-grown tools and technologies.
- The Connector Catalog provides connectors to popular third party technologies such as metadata repositories, databases, data formats and platforms; data movement engines, data virtualization engines, dev ops tools, analytics/AI tools, data catalogs, MDM and user directories, CMDBs, SDLC tools, ...
- The Content Packs provide metadata, reference data and connector configurations to help get you up-and running with the supplied capability of Egeria. They are formatted as Open Metadata Archives and can be configured to load at server startup using the Administration Services or while the server is running using the Runtime Manager OMVS.
Developer platform¶
Provides frameworks, APIs, and hosting platforms for building an integrated metadata and distributed governance solutions.
The developer platform contains the core Egeria implementation and provides support for integrating third party technology into the open metadata ecosystem and extending Egeria to run in different environments or to use different infrastructure services.
Its use is described in the developer's guide.
- There are clients written in both Java and Python to aid programmers calling the Egeria services.
- The utilities and reports support common tasks that retrieve and update metadata.
- The Unit Test Utilities help to unit test Egeria specific components. They work with standard test harnesses.
- The Function Verification Tests are used to test Egeria's runtime. However they illustrate how to automate tests for your components that work with the Egeria runtime.
-
The Conformance Test Suite (CTS) supports the testing of third party connectors. Each type of connector or service is supported by its own test workbench.
-
Open Metadata and Governance (OMAG) registered services are dynamically loaded in the OMAG Server Platform. This means they can be added and removed as needed to create a customized platform. This may include registered services written by the Egeria community and supplied by third parties.
- Access services provide specialist APIs / events for different types of tools. They work with the pre-defined open metadata types and use the repository services to access metadata.
- Engine services provide the services that host a specific type of governance engine. The governance engines collectively provide active governance to the assets and their associated metadata.
- Integration services each provide a specialized API to integration connectors. These are hosted in an integration daemon. The purpose of the integration services is to simplify the implementation and management of connectors that integrate metadata exchange with third party technologies.
- View services provide the services used by UIs. They are typically fine-grained services and they run in the view server. The use of the separate server (and server platform) enables an extra firewall to be erected between the view servers and the metadata servers and governance servers, hiding the internal systems from end users.
-
The Open Metadata Types provide definitions for the different types of metadata needed by an organization. The open metadata type system is extendable; however, by providing a comprehensive starter set, and encouraging tools to use them, Egeria ensures metadata can be seamlessly shared amongst them.
-
The Framework Services provide Egeria clients to support metadata retrieval for connectors defined by the frameworks.
- Connected Asset supports the creation of connectors and the retrieval of metadata about the attached asset.
- Open Metadata Store supports the retrieval and maintenance of any type of open metadata.
- Open Integration Service provides the runtime services for integration connectors.
- Open Governance Service supports the execution of engine actions and governance action processes.
-
The OMAG Server Platform provides a multi-tenant runtime platform for OMAG Servers. Each OMAG Server hosts the connectors along with the Egeria services to integrate third party technology.
- The platform chassis uses Spring Boot to provide the web server and REST API support for the platform.
- The server chassis uses Spring Boot to provide the web server and REST API support for a single OMAG Server.
- The administration services supports configuring the OMAG Platform and Servers. Details of how to use the admin services are provided in the administration guide
- The server operations and platform services provide the means to start, stop and query the OMAG Servers and services running on an OMAG Server Platform.
- The multi-tenancy management module supports multiple OMAG Servers running on an OMAG Server Platform.
- The repository services provide the basic ability to share metadata between metadata repositories. The metadata repositories are organized into open metadata repository cohorts. These cohorts define the scope of the metadata sharing and ensure metadata is available to all consumers within the cohort.
- The metadata security module provides customizable authorization checks for calls to the OMAG Server Platform, OMAG Server and the open metadata instances themselves.
- A governance server makes use of open metadata to actively manage an aspect of the digital landscape. The governance server services each provide the principle subsystem of a type of governance server.
- The generic handlers provide support for the type specific maintenance and retrieval of metadata that follows the open metadata types. This includes managing visibility of metadata through the Governance Zones, calls to Open Metadata Security and metadata management using templates.
-
The open metadata frameworks define the interfaces implemented by components that "plug-in" to Egeria, either to integrate calls to third party technology or extend the function of Egeria. The frameworks are as follows:
- Audit Log Framework (ALF) - extensions for all types of connectors to enable natural language diagnostics such as exceptions and audit log messages.
- Open Metadata Framework (OMF) - base framework for open metadata definitions.
- Open Connector Framework (OCF) - base framework for all types of plug-in components called connectors.
- Open Integration Framework (OIF) - specialized connectors for metadata exchange and synchronization with third party technologies.
- Survey Action Framework (SAF) - specialized connectors called survey action services that support automated metadata discovery.
- Governance Action Framework (GAF) - specialized connectors for the triage and remediation of issues found in the digital landscape.
- Context Event Framework (CEF) - specialized connectors for context event management.
Deployment runtimes¶
The runtimes package up the services to simplify the process of deploying Egeria's capability.
There are two runtimes:
- OMAG Server Platform provides support for dynamically configuring and running multiple OMAG Servers.
- OMAG Server runtime provides support for running a single OMAG Server. It is designed for static production deployment into a cloud-native Kubernetes environment.
Deployment resources¶
Aim to simplify the process of deploying the OMAG Server Platform and its connectors into an operational environment.
- The Egeria docker image is built daily and pushed to Quay.io. It contains an OMAG Server Platform. You can download it and use it in your own container environments. It is used by the helm charts.
- The Kubernetes Helm charts make use of the docker image to create a rich Egeria deployment used in the open metadata labs.
- The Kubernetes operators are in development. They will provide an easy way to control an Egeria deployment running on Kubernetes.
Support for docker compose was removed in release 3.5.
Understanding the roadmap¶
Current status¶
Following is an overview of the content status of the functions in Egeria's latest release (5.3).
As you can see, some progress has been made on all layers. However, since they do build on one another, most of the early work has been focused on establishing the frameworks, connector APIs and other services to provide the developer platform. The developer platform provides the libraries and interfaces to build connectors to integrate third party tools along with the runtime to host these connectors and manage the metadata exchange.
Today we have a robust OMAG Server Platform and the ability to configure OMAG Servers that host specific types of connectors to third party tools. The continual development of the registered services is broadening the types of metadata, and tools, that can integrate with Egeria through the federated query mechanism, supported by the repository services. With this expansion, more solutions are envisaged, along with a runtime environment that can be downloaded and used immediately. Many features are available through Egeria's command line interface, or through simple python scripting. Thus, a team can be up and running quickly. The dynamic configuration and Java frameworks that support new connectors and services development are still there and can be used to further expand Egeria's capabilities.
History¶
The initial implementation of Egeria focused on the Open Metadata Repository Services (OMRS) to support different types of metadata repositories exchanging metadata via an open metadata repository cohort. This also involved the build out of the OMAG Server Platform to host this code. The aim was to demonstrate how third party metadata servers could exchange metadata. This capability was delivered along with two repository connectors for the following third party connectors along with the Conformance Test Suite.
Through 2020, our focus shifted to the integration platform as we added connector implementations for popular third party technologies (see the connector catalog) and built out the ecosystem user interface (UI) that enables an organization to:
- configure OMAG Servers on OMAG Server Platforms
- visualize the open metadata types through the type explorer (TEX)
- visualize open metadata instances in a single repository or across the open metadata repository cohorts that a server is connected to.
- visualize to cohort and query the operational status of the OMAG Servers and services operating in the open metadata ecosystem
- configure OMAG Servers and deploy them to OMAG Server Platforms
The ecosystem UI makes calls to specialized REST services supported by a type of OMAG Server called the view server. The view server is new for 2020 and enables the REST APIs to the UIs to be deployed in a DMZ and the metadata servers to be behind an additional firewall. It also takes much of the load for supporting end users off of the metadata servers.
In 2020 support for a new type of OMAG Server called the integration daemon was also added. This server supports integration services that can host integration connectors dedicated to exchanging metadata with specific third party technologies.
2021 had a focus on governing metadata. There is a new OMAG Server called the engine host that runs governance engines. These are supported by new access services for governance.
This new website was added to the project in 2021, and it has resulted in more interest in consuming Egeria.
2022 continued the focus on metadata governance. The following OMASs were refactored to call the generic handlers rather than direct calls to the repository handler.
- Data Engine OMAS
- Asset Catalog OMAS
- Asset Lineage OMAS
There was investment in both the function and performance of the generic handlers, which provide many of the metadata governance functions supported by all OMASs, such as metadata security, provenance validation, anchor management, LatestChange classifications, effectivity dating, memento management and de-duplicating query results.
Integration with third party technologies made good progress with the addition of OpenLineage support, the new JDBC, Hive Metastore, schema registry, OpenAPI Specification and Apache Kafka connectors.
2023 and 2024 saw a dramatic change in the project's direction as it shifted its primary focus from providing libraries for tools vendors to supporting organizations that are looking to create an open metadata ecosystem. This resulted in more samples, view services, content packs and a new range of UIs. The OMAG Server Platform has had some internal restructure and clean up of deprecated code. The result is a simpler coding experience for new community members.
Future Plans¶
After the release of version 5.0, the team continues to focus on usability, simplification and publicity around the new platform and its advanced capability. There will also be more connectors, increasing Egeria's reach across the IT landscape. Functionally, you will see the completion of the Context Event Framework (CEF) and its supporting services, plus more view services will emerge as the governance solutions mature. The speed of this rollout depends on our ability to rebuild the community.
How Egeria has evolved over the years. This reflects that the scope has grown from a desire to provide embeddable integration libraries for use in data tools and platforms, to today's desire to directly support teams in their governance, data and AI projects. The libraries and frameworks are still there. However, most users will experience as a running services, preloaded with metadata to accelerate their adoption.
Raise an issue or comment below