News

Oracle Updates Autonomous Data Warehouse

• By John K. Waters
• March 18, 2021

Oracle showed off a new set of enhancements to its Autonomous Data Warehouse this week aimed at data analysts, data scientists, and line-of-business users. With this update, the company is providing a single data platform built to allow those users to ingest, transform, store, and govern their data to run diverse analytical workloads from any source, including departmental systems, enterprise data warehouses, and data lakes.

Andy Mendelson, EVP of Oracle's Database Server Technologies Group, made the announcement during an Oracle Live event on Wednesday. This release is "expanding the vision" of the Autonomous Data Warehouse, he said.

"Today, anybody can go to the Autonomous Data Warehouse with the push of a button in a couple of minutes," he said. "We also want to make it easy for anyone to get their data into the warehouse and get value out of it. With this next generation, we provide a set of easy-to-use, no-code tools that uniquely empower business analysts to be citizen data scientists, data engineers, and developers."

The list of enhancements in this release includes, among others:

• Built-in Data Tools: A simple, self-service environment for loading data and making it available to extended teams for collaboration. Users can load and transform data from their laptops or the cloud by dragging and dropping. They can then automatically generate business models; quickly discover anomalies, outliers, and hidden patterns in their data; and understand data dependencies and the impact of changes, the company says.
• Oracle Machine Learning AutoML UI: By automating time-intensive steps in the creation of machine learning models, the AutoML UI provides a no-code user interface for automated machine learning to increase data scientist productivity, improve model quality, and enable even non-experts to leverage machine learning, the company says.
• Oracle Machine Learning for Python: Data scientists and other Python users can use Python to apply machine learning on their data warehouse data and leverage the high-performance, parallel capabilities and 30+ native machine learning algorithms of Oracle Autonomous Data Warehouse.
• Oracle Machine Learning Services: DevOps and data science teams can deploy and manage native in-database models and ONNX-format classification and regression models outside Oracle Autonomous Data Warehouse. They can also invoke cognitive text analytics. App developers have easy-to-integrate REST endpoints for all functionality, the company says.
• Property Graph Support: Graphs help to model and analyze relationships among entities (for example, a social network graph). Users can now create graphs within their data warehouse, query graphs using property graph query language (PGQL), and analyze graphs with more than 60 in-memory graph analytics algorithms.
• Graph Studio UI: Builds on property graph capabilities of the autonomous warehouse to make graph analytics easier for beginners, the company says. It includes automated creation of graph models, notebooks, integrated visualization, and pre-built workflows for different use cases.
• Seamless Access to Data Lakes: This update extends the ability of the autonomous warehouse to query data in Oracle Cloud Infrastructure (OCI) Object Storage and popular cloud object stores with three new data lake capabilities: easy querying of data in Oracle Big Data Service (Hadoop); integration with OCI Data Catalog to simplify and automate data discovery in object storage; and scale-out processing to accelerate queries of large data sets in object storage.

The Autonomous Data Warehouse is winning fans internationally, says IDC analyst Carl Olofson. "Our research, based on interviews with several customers around the globe, shows that those Oracle Autonomous Data Warehouse customers have achieved approximately 63 percent reduced total cost of operations, while increasing the productivity of data analytics teams by 27 percent, with breakeven on their investment having occurred in an average of five months," Olofson said in a statement. "This ROI included significant productivity gains across data, analytics, and developer teams. While individual customer results may vary, the benefits found in this study are indicative of the kind of improvements that most may expect. With these new intuitive integrated tools incorporated in Oracle Autonomous Data Warehouse, it is reasonable to expect that productivity gains will further increase, enabling businesses to achieve an even better ROI."