Oracle Announces New ML Capabilities for MySQL HeatWave

(Blackboard/Shutterstock)

Today, Oracle announced its MySQL HeatWave database service now supports in-database machine learning as an additional capability to the service’s existing transaction processing and analytics. HeatWave ML is complimentary for MySQL Heatwave customers in all OCI regions and is available today.

According to Oracle, “HeatWave ML fully automates the ML lifecycle and stores all trained models inside the MySQL database, eliminating the need to move them to a machine learning tool or service using data movers and connectors.” The company says the integration of new ML capabilities with existing functionality will allow customers to use real-time data with less complexity, latency, and cost of ETL duplication.

Through automation of the ML lifecycle in-database, including pre-processing of data, algorithm selection, intelligent data sampling, feature selection, and hyper-parameter tuning, Oracle affirms that HeatWave ML allows customers to save time and cost with ML initiatives without the need for data scientists. The company also asserts that “all models trained by HeatWave ML are explainable, improving regulatory compliance, fairness, repeatability, causality, and increasing trust in machine learning.”

To complement this announcement, Oracle published machine learning benchmarks performed across publicly available classification and regression datasets. One claim the company makes is that, compared to Amazon’s Redshift ML, Heatwave ML trains models 25x faster at 1% of the cost, and it scales with cluster size.

“Just as we integrated analytics and transactional processing within a single database, we are now bringing machine learning inside MySQL HeatWave,” said Edward Screven, chief corporate architect, Oracle. “An increasing number of customers have migrated from Amazon and other cloud database services to MySQL HeatWave because of the significant performance gains and lower costs.”

Click to enlarge. Source: Oracle

Oracle claims its 10TB TPC-DS benchmark results show that other cloud services cost more and are slower than MySQL Heatwave (see table at right). The company declares that Snowflake is 14.4x slower, Redshift is 4.8x slower, Azure Synapse is 14.9x slower, and Google BigQuery is 12.9x slower, with each service being more expensive. Oracle has publicly provided the benchmarking code for ML and TPC-DS and customers can run the benchmarks by visiting this link.

Oracle also announced new features for MySQL HeatWave, including real-time elasticity that allows scaling of HeatWave clusters up or down to any number of nodes with no downtime or need for manually rebalancing the cluster. The company also announced data compression, a feature it says enables processing of twice the data per node at half the cost. There is also a new pause-and-resume function to allow customers to pause HeatWave for cost savings, with the added capability of automatic reloading of the required MySQL Autopilot data and statistics upon resumption.

Source: Oracle

Oracle says its customers are finding value in HeatWave ML and MySQL HeatWave’s new features, including Astute Business Solutions, an Oracle PartnerNetwork software provider. “We recently had an opportunity to use the machine learning capabilities of HeatWave ML. We found it very innovative, easy-to-use, very fast and most important it is secure since the data or the model don’t leave the database,” said Arvind Rajan, co-founder and CEO of Astute Business Solutions. “We believe that providing in-database machine learning is of significant interest to our clients and will further accelerate the adoption of MySQL HeatWave.”

Another client, VRGlass, a Brazilian SaaS producer of metaverse apps and equipment, saw increased database functionality: “Motivated by the progress achieved within the Oracle for Startup program, VRGlass migrated all application data to MySQL HeatWave from AWS EC2. Within three hours, we achieved a 5x increase in database performance for a virtual event that accommodated more than 1 million visitors and 1.7 million sessions with greater security and at the half the cost,” said the company’s CEO, Ohmar Tacla.

With today’s announcements, the company is adding to its MySQL open source cloud database capabilities, especially in the realm of machine learning.

“Oracle announced MySQL HeatWave with Autopilot last August, which may very well have been the single greatest innovation in open source cloud databases in the last 20 years to that point,” said Carl Olofson, research vice president, data management software, IDC. “Now Oracle has gone beyond its original unifying of OLTP and OLAP in HeatWave, with MySQL HeatWave ML. Oracle is bringing all of the machine learning processing and models inside the database, so that customers not only avoid managing ML databases apart from the core database, but also eliminate the hassles of ETL, gaining speed, accuracy, and cost-effectiveness in the bargain.”

To learn more about HeatWave ML, read the technical brief here.

Related Items:

Database Migrations Shift Into High Gear

O’Reilly Report Shows Relational DB Momentum, No AI Slowdown

Who’s Winning the Cloud Database War