When Oracle introduced Oracle HeatWave in May 2021, Larry Ellison, its Founder and CTO said, “It is the physical manifestation of nearly 10 years of deep database engineering techniques, and over 5 dozen patents. It demonstrates what real cloud database innovation looks like in 2021.”
What’s HeatWave, and why did it get so much press? Oracle HeatWave lets its users run Online Transaction Processing (OLTP) and Online Analytical Processing (OLAP) workloads directly from MySQL databases. Why is it such a big deal? Indeed, in a way it is. It allows users to bypass the ETL procedures to get transactional data into another database for analytics.
Also, Why MySQL? Now, that’s quite clearly evident. As of January 2022, according to the data published by Statista, MySQL was the most popular open source database management system (DBMS) in the world, with a ranking score of 1206. PostgreSQL, with a score of 607 and MongoDB, with a score of 489 were a distant 2nd and 3rd respectively.
HeatWave, if we go by the benchmarking done by Oracle, “increases MySQL performance by orders of magnitude for analytics and mixed workloads, without any changes to existing applications.”
Oracle also boasts of massive cost savings when compared with AWS Redshift and Snowflake. “ HeatWave is 18x faster, provides 110x better throughput and is 2.4x cheaper than AWS Aurora – managed relational database engine by AWS for OLAP queries resulting in 42x better price performance.
HeatWave, since its launch last year, has been received well by its customers. After all in a market that’s already too cluttered, HeatWave offered a solid differentiation. By incorporating the hardware and software optimisation and adding analytics to it, Oracle found the sweet spot.
Just when March was about to end, Oracle made yet another big media splash that the MySQL HeatWave now supports in-database machine learning (ML). This additional capability provides users a new MySQL cloud database service solution. HeatWave ML is included with the MySQL HeatWave database cloud service in all 37 Oracle Cloud Infrastructure (OCI) regions. This announcement represents the 3rd major release of MySQL HeatWave in about 15 months.
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. HeatWave ML will be available to its users at no additional charge.
Oracle claims that no other cloud database vendor or open-source database provides such advanced ML capabilities inside a database. For the first time in the industry, Oracle also published ML benchmarks performed across a large number of publicly available classification and regression datasets. On average, HeatWave ML trains models 25 times faster, at 1 percent of the cost of Redshift ML, and scales with cluster size. MySQL HeatWave customers can now train models more often and keep them updated for increased prediction accuracy. New benchmark results reinforce MySQL Heatwave superior price/performance.
Industry standard 10TB TPC-DS* benchmark results prove how other cloud services are slower and more expensive than MySQL HeatWave:
· Snowflake: 14.4 times slower and more expensive
· Amazon Redshift: 4.8 times slower and more expensive
· Azure Synapse: 14.9 times slower and more expensive
· Google BigQuery: 12.9 time slower and more expensive
In addition to MySQL HeatWave ML, Oracle also released additional innovations, including real-time elasticity, which enables customers to upsize and downsize their HeatWave cluster to any number of nodes, without any downtime or read-only time. Also included is Data compression, which enables customers to process 2 times the data per node and lowers costs by nearly 50 percent, while maintaining the same price performance ratio. And a new Pause-and-resume function, which enables customers to pause HeatWave to save costs. Upon resumption, both the data and the statistics needed for MySQL Autopilot are automatically reloaded into HeatWave. The new features introduced in the latest MySQL HeatWave release are available now on OCI across all 37 Regions.
After the recent launch, dynamicCIO got an exclusive opportunity to speak to Nipun Agarwal, Senior VP, MySQL HeatWave Development, Oracle Corp. We spoke on a wide variety of issues including the capabilities that MySQL HeatWave ML unleashes and the benchmarks that show it is ahead of all of its competitors. Listen in…