Oracle OpenWorld 2016 - Data Integration Recap

I know it's been about a month since Oracle OpenWorld 2016 concluded, but I wanted to give a brief recap on a few things that I thought were interesting in the data integration space. During the week prior to OpenWorld, I had the privilege to attend the Oracle ACE Director Briefing. Over 2 days, ACE Directors were provided an early preview of what's to come down the Oracle product pipeline. The importance of the event is easy to note as Thomas Kurian himself spends an hour plus providing the initial product overview and answering questions. The caveat, the entire session is under NDA (as you might imagine). But, the good thing is that several of the early preview products were announced the following week at Oracle OpenWorld. Here's what I saw that might impact the Oracle Data Integration product area most.

Data Flow ML

Take an ETL tool, add the cloud, and mix in the latest Big Data technologies and methodologies and you have Data Flow ML. This cloud-based tool is built for stream or batch processing, or both, following the Lambda architecture. Data is ingested into Kafka topics, transformed using Spark Streaming, and loaded into a final target, which may be created automatically by DFML. Along the way, Spark ML is used to profile the data and make suggestions for how to enrich the data with internal or external sources. The technology is still in its early stages but keep an eye out on the Rittman Mead blog for more information over the next few months.

Data Integration in the Cloud

Oracle Data Integrator Cloud Service is coming soon and with it, new licensing options. ODI can be deployed in the cloud on Java Cloud Service or Big Data Cloud Service, or it can be deployed on-premises for more of a hybrid environment. From a licensing perspective, ODICS can be a monthly subscription or you can BYOL (bring your own license) and run ODI from any compute resource. This flexibility allows you to pushdown the transformation execution to the location of the data, rather than moving the data to the transformation engine - a standard for Oracle Data Integrator.

Oracle Data Integrator 12.2.1.2

Coming soon, the next patchset release for Oracle Data Integrator 12c. Features discussed at Oracle OpenWorld were:

• Git Integration and Smart Merge:
  This release will introduce a new integration for lifecycle management, Git, adding to the current integration option of Subversion. Not only that, but ODI will finally provide "smart" merge functionality to allow an automated merge of a branch into the trunk.
• Kafka and Spark Streaming:
  With ODI for Big Data, streaming integration is coming. Use of Apache Kafka as a source or target and Spark Streaming integration for transformations will allow for more real-time processing of data. The introduction of Cassandra as a data source is another enhancement for the Big Data release.
• RESTful Connectivity:
  Another long awaited feature is REST web service integration. A new technology, similar to the SOAP web service integration, will be available and able to connect to any RESTful service. Along with that, BICS and Storage Cloud Service integration will be introduced.

There are definitely many other interesting looking products and product updates coming (or already here), including GoldenGate Service Architecture, updates to the GoldenGate Cloud Service, Big Data Cloud Service, Big Data Compute and several others. It’s an interesting time as the Oracle shift to the cloud continues - and data integration won’t be left behind.