Another amazing year is in the books, and yet again I cannot believe the pace at which the GoldenGate team are delivering exciting new functionality that goes boldly into new areas while also staying laser-focused on solving real-world customer needs. After the huge surge we did for Cloud World 2024 (GoldenGate 23ai, Multicloud, Duality with Data Streams, ZeroETL, Azure and Solace), I did not think we could top it for 2025 and the team proved me wrong again!

This year is actually the 30-year anniversary of GoldenGate, which was originally founded in San Francisco in 1995. For three decades this technology has been in perpetual “growth mode” and it is only fitting that 2025 would be the biggest year yet. Personally, I think the reason this technology continues to dominate is that founding technical team remains in place after all this time.  GoldenGate’s original Development leadership, Lead Architect, first Product Manager, first Customer Success Manager and Partner leadership are still all part of this team. In an era where everything is shifting to AI, it is still the people that matter most for creating and sustaining world-class enterprise software.

The GoldenGate platform is focused on providing solutions for real-time data fabric, especially in mission-critical enterprise systems. These modern solutions can provide multicloud data integration with real-time change data capture, data replication, streaming data pipelines along with stream analytics. GoldenGate can work in on-premises environments, Oracle Cloud, AWS, Azure Google Cloud Platform and even running in semi-disconnected architectures for IoT and device-centric computing. GoldenGate is a modern microservices architecture that is well supported in Docker, Kubernetes and OpenShift.

One of the reasons GoldenGate remains so popular after 30yrs is the ongoing investments in supporting the full set of Applications, Databases, Data Lakes, Vector DBs, NoSQL and Streaming technologies that are popular across the software industry. Most of GoldenGate’s larger customers standardize on the technology as their enterprise solution.

This year’s Oracle AI World 2025 continues our long tradition of rolling out exciting new features, enhancements, partnerships and capabilities for the GoldenGate platform. Without further ado, here are my Top 7 picks for this year!

1.    GoldenGate 26ai – AI centric release update with exciting new features
2.    Stream Analytics 26ai – AI centric major release for our streaming customers
3.    New AI Certifications – for Oracle AI Data Platform and Oracle Autonomous AI Lakehouse
4.    Azure General Availability – for OCI GoldenGate running in Azure cloud
5.    Container Images – run “GG as a Service” in your own local data center!
6.    Veridata 26c – available on-premises and now as part of OCI GoldenGate
7.    Air-gap Replication for Secure Networks – for best-in-class high security

Bonus announcement!

8.    All new Learning Paths and Professional Certifications – for GG developers

Let’s take a closer look at each of these…

1. GoldenGate 26ai

Obviously, the star of the GoldenGate announcements this year is our new 26ai release update. This is a part of our long-term release strategy and the main branch we will be asking all customers to move to, it is a continuation of our most stable 23.x branch but with a whole host of exciting new AI and Cloud capabilities.

Here’s a bullet point summary of what you can expect to see in 26ai:

• Oracle AI Database 26ai – full certification and support for all HA/MAA patterns
• Oracle AI Data Platform – full certification for real-time integration
• Oracle Autonomous AI Lakehouse – full certification for real-time integration
• OCI AI Generative AI Agents – full support for GoldenGate tools as part reactive AI Agents
• Oracle Integration Cloud AI Agent Framework – support for adding GoldenGate Data Streams as integration tools
• GoldenGate AI Microservice – standard option in GoldenGate Deployments (can work with local or remote LLMs to bring GenAI services to real-time replication use cases)
• Air-gap Data Replication – for maximum security isolated networks
• Full container downloads – run “GoldenGate as a Service” in your own infrastructure (eg; Docker, K8S, OpenShift etc)
• Auto-schema – for simplified schema changes in downstream targets
• New 3rd Party Sources/Targets – YugabyteDB, Microsoft Fabric Event Stream, EnterpriseDB, Google Spanner and many more for Veridata 26c

This 26ai release (targeted for January 2026 as an RU/BP update) is a major milestone for the GoldenGate pivot to next-generation AI capabilities.  Our vision for the future is a GoldenGate platform that solves the hardest multicloud Data Fabric problems in the simplest and most cost-effective ways. We will be using Generative AI to help our customers make their data AI-ready, simplify how they monitor and triage GoldenGate and use natural language to create and manage their GoldenGate ecosystems.

Indeed, there has never been a better time to be on the GoldenGate platform!  Read-on with Alex Lima’s blog to learn more about the GoldenGate 26ai updates.

2. Stream Analytics 26ai

A major aspect of GoldenGate growth and vision pivot around stream processing and stream analytics, this is where we can do stateful processing of real-time data in rolling and tumbling time-windows. This allows us to tackle real-time data preparation, streaming ETL and time-series analytics from the data pipelines. Our GoldenGate Stream Analytics provides these foundation capabilities for stream processing and time-series analytics. Stream Analytics works natively with GoldenGate’s replication capabilities and empowers customers to take data events from the data layer and process those events seamlessly within exact-once, ordered and transactionally-consistent streaming pipelines.

The next generation Stream Analytics 26ai capabilities include a host of new functionality designed to make it simpler and more intuitive to use AI & ML within streaming pipelines, here’s a summary:

• Data Streams Integration – directly consume GoldenGate events, bypassing complex Kafka clusters
• Modern UX – updated to the latest Oracle Redwood frameworks for a beautiful low-code experience
• AutoML with Oracle AI Database 26ai – deeply integrated with the newest Oracle AI Database
• LLM/GPT Operators – use AI for data enrichment, data quality in real-time pipelines
• Vector Embeddings – as part of streaming pipelines, then write vectors out to analytic data stores

With these advancements in real-time AI and ML capabilities and an easy-to-use low-code editor, we are working hard to bring the benefits of a streaming architecture to everyone!  Don’t miss Alex Kotopoulis’ detailed blog with even more examples of what you can now do with Stream Analytics 26ai.

3. New AI Certifications

Well, I guess that I have already ‘let the cat out of the bag’ with my #1 favorite announcement (26ai), but the sheer number of new certifications and supported AI technologies is worthy of its own section! The GoldenGate team did an enormous amount of work to align with other Oracle teams across the whole company to make sure that we are providing a unified release cadence that provides best-in-class support for Oracle technologies that depend on GoldenGate for real-time data feeds.  Here’s a breakdown of key new AI related technologies that we are announcing as certified and/or supported integrations:

• Oracle AI Database 26ai – full certification and support for all HA/MAA patterns
• Oracle AI Data Platform – full certification for real-time integration
• Oracle Autonomous AI Lakehouse – full certification for real-time integration
• OCI AI Generative AI Agents – full support for GoldenGate tools as part reactive AI Agents
• Oracle Integration Cloud AI Agent Framework – support for adding GoldenGate Data Streams as integration tools

And, further… we are also adding support for the Oracle Cloud Infrastructure Generative AI, OpenAI, Azure OpenAI and custom APIs for local and remote LLMs that can be used directly within GoldenGate 26ai itself. That means that some of the most popular OpenAI models (eg; GPT-4 Turbo, GPT-4 Turbo with Vision, GPT-3.5-Turbo, text-davinci-003, ada, babbage, curie, davinci, text-embedding-3-large, text-embedding-3-small, text-embedding-ada-002) as well as LLMs and embedding models from any compatible tool layers.

Check out this post by Deniz Sendil to learn more about the Oracle AI Data Platform and Oracle Autonomous AI Database integrations with GoldenGate!

4. Azure General Availability

Last year at Oracle Cloud World 2024, we pre-announced our multicloud partnerships with Microsoft and Google.  This year we are thrilled to announce General Availability of GoldenGate@Azure! This is a fully managed service, essentially the same OCI GoldenGate service but running in Azure data centers and commercially available with Microsoft Azure Consumption Commitments (MACC). GoldenGate on Azure can of course be used to integrated with Oracle Database@Azure. The service can also be used to integrate with the full suite of Azure services including Fabric, Fabric Open Mirroring, Fabric Event Stream, AzureSQL Databases, Synapse, CosmosDB and even Snowflake or Databricks running in Azure.

This is a great time to be an Azure customer because now you can have access to the #1 real-time data fabric GoldenGate platform. Regardless of whether Azure customers are running Oracle databases, Postgres databases, SQL Server or anything else; Azure customers can now create real-time data pipelines that connect data events across all their most important Azure assets and connect those data events to Oracle, Google and Amazon cloud services.

One very important observation that I want to highlight here is that this new multicloud integration is far superior to ordinary Azure Marketplace integrations. This GoldenGate@Azure is a fully-managed service, with console-to-console integrations and a deep engineering partnership across Azure and Oracle GoldenGate development organizations. Unlike self-managed VMs on Marketplace, the new GoldenGate capabilities are part of next-generation multicloud system architectures that are beneficial for our customers.

Check out Nick Wagner’s blog post for more details about how to use the service and which regions are available now!

5. Container Images

In this modern world of cloud-native computing it is easy to forget that most of the world’s largest companies continue to run vast IT estates in on-premises servers or as self-managed software running in cloud compute. There are various economic, technical and security reasons that some data continues to run on-premises or in self-managed software. Oracle is the only hyperscale cloud vendor that provides a real-time data fabric solution that can run on-premises, in self-managed cloud infrastructure, as cloud-native services and in a semi-disconnected device architecture. For these on-premises and self-managed use cases, our largest customers are sometimes managing 100’s or 1000’s of GoldenGate deployments. These customers need the benefits of running “GoldenGate as a Service” but within their own on-premises and self-managed infrastructure; that is why we are introducing a whole new suite of pre-built GoldenGate software containers designed to simplify CICD automation for self-managed GoldenGate.

So, go ahead check out the Oracle Container Registry GoldenGate Repository and try these yourself!  You’ll find:

• Complete builds – configured, ready to run
• Latest bundle patches – stay up to date
• Built-in best practices – 
  • No more lengthy installations or in-place patching
  • Simple docker pull command
  • Upgrade GoldenGate in seconds
  • Run with Kubernetes or OpenShift pods 
  • Automate from AWS, Azure or Google
  • Build your own “GoldenGate as a Service”
• All popular packages – 
  • GoldenGate for Oracle, MySQL, PostgreSQL, SQL Server and GoldenGate for Distributed Applications & Analytics (Snowflake, Databricks, BigQuery, Microsoft Fabric, Kafka etc.)

Also, you can check out Alex Lima’s new blog post that gives detailed instructions on how to get started with these containers right now.

6. Veridata 26c

One of the top analysts in the software industry recently told me that he thinks that Veridata is one of the “hidden gems” of the Oracle portfolio – and I agree! GoldenGate Veridata is an important tool that works with or without other GoldenGate components and it also works with or without Oracle Databases.

Veridata brings best of breed tooling for:

• Data Governance – for mission-critical systems to guarantee that there is no missing data
• Compare and Repair – compare all records between two different data sets; then automate repair or data
• Heterogeneous – run Veridata jobs on Oracle or Non-Oracle data stores

New capabilities in Veridata 26c include:

• MariaDB and SingleStore
• MongoDB, Atlas, AJD, JCT, DocumentDB and Azure CosmosDB
• Databricks (Delta Tables)
• Built-in UX for job scheduler and server config
• LDAP/Active Directory support
• Pre-built OCR image

And…perhaps most importantly, we are now offering Veridata as a fully-managed deployment within OCI GoldenGate. This brings next-level ease-of-use with cloud-native automation for all our Veridata customers.  And, “yes!” this is also going to be available in all the GoldenGate multicloud deployments for Azure and Google.

For more information about running Veridata in OCI GoldenGate, don’t miss Julien Testut’s recent post on all the new capabilities we’ve launched with the OCI GoldenGate service!

7. Air Gap Replication

In this complex and modern world of cloud computing that we all live in now, it is crucial to be able to have high security options that can physically guarantee the isolation of certain kinds of highly privileged information. That’s one of the main reasons why “air-gapped networks” exist; to be sure that some protected assets have no contact with any other unsecured networks. Typically we would see these kind of air-gap networks in government organizations but increasingly they are also becoming more commonplace for sensitive commercial organizations as well.

Our new GoldenGate feature supports the encoding and decoding of data events (eg; DML and DDL transactions), including large object support (LOBs). Technically speaking, this enables a flow of transactions through the physical data diodes (usually called “Guards”) that can inspect and protect data that needs to move through the air-gap.  The GoldenGate development team had an innovative partnership with Oracle Government, Defense and Intelligence (GDI) leadership to make certain that this feature is ready for the strictest, real-world demands that the “three letter agencies” may place on it.

If you’d like to know more about this capability, be sure to read the new blog post written by Denis Gray and Nick Ratanasin that covers more in-depth information.

Ok, so those of you reading this post who know me already knew that there was no way I could contain myself to ‘only’ 7 top items… so here it is, your #8 bonus top announcement!

8. Learning Paths and Professional Certifications

Now that we have your attention about all the cool new announcements for GoldenGate, are you curious about making the most of it? There’s never been a better time to dive in, with a whole menu of hands-on workshops and learning options available—no matter where you are on your data integration journey. From in-person, interactive workshops right at your office to engaging online sessions, there’s something for everyone, whether you’re a seasoned GoldenGate veteran or just starting out.

The in-person workshops are a great way to learn as a team and get all your burning questions answered by experts—think of it as a data integration boot camp, but with friendly faces and real-life scenarios. You’ll get a taste of everything, from understanding the basics of the GoldenGate platform to exploring hot topics like event-driven data, AI and machine learning, multicloud architectures, and more. Plus, there are hands-on technical deep dives for those who want to roll up their sleeves and get nerdy.

Prefer to learn from the comfort of your desk or sofa? Check out Oracle LiveLabs, where you can jump into free online workshops and labs whenever you want. These interactive sessions let you experiment with new GoldenGate features or build pipelines at your own pace—perfect for anyone who likes to learn by doing.

For those looking to level up with formal training, Oracle University has your back. Their mix of free courses, detailed fundamentals, and upcoming advanced topics means you can build your expertise step-by-step. And if you want to showcase your GoldenGate chops, there are certifications available (and more coming soon)! No matter your learning style, it’s easy—and fun—to get started and go further with GoldenGate.

You can check out all the GoldenGate specific workshops, labs and training in our new GoldenGate Knowledge Hub – and also be sure to take a read on Pete Inzana’s blog post about all the new and upcoming learning paths that are available to you now!

Concluding thoughts about AI World 2025 and GoldenGate

Well if you’ve made it this far you can clearly see that our theme for this year is AI and I hope I’ve convinced you that we have knocked a “home run” on the new feature announcements. It has been another great year (our 30th anniversary!) for GoldenGate at Oracle AI World 2025. I for one believe that the whole GoldenGate organization has done a fantastic job bringing all these features to you, our customers.

I mentioned this at the top of the blog and I want to conclude by reflecting on this once again… in a world of AI and IT we cannot forget that it is still ultimately the hard work of people, more specifically teams of people who ultimately bring forward these innovations and make them real.  Cheers to our customers who make it all worthwhile and to the vast team of GoldenGate collaborators at Oracle who bring these innovations to market!