Oracle has announced that MySQL HeatWave is available on Amazon Web Services (AWS). MySQL HeatWave is the only service that combines OLTP, analytics, machine learning, and machine learning-based automation within a single MySQL database.
AWS users can now run transaction processing, analytics, and machine learning workloads in one service, without requiring time-consuming ETL duplication between separate databases such as Amazon Aurora for transaction processing and Amazon Redshift or Snowflake on AWS for analytics and SageMaker for machine learning.
“Oracle believes in giving customers a choice. Many of our MySQL HeatWave customers migrated from AWS. Others wish to continue running parts of their application on AWS. Those customers face serious challenges including exorbitant data egress fees charged by AWS and higher latency when accessing a database service running in Oracle’s cloud,” said Edward Screven, chief corporate architect, Oracle.
“We are addressing these issues while delivering outstanding performance and price performance across transaction, analytics, and machine learning compared to other database cloud providers—even Amazon’s own databases running on AWS, where you’d think they would have an advantage. We wanted to offer AWS customers this choice to benefit from MySQL HeatWave innovation without moving their data from AWS, or developers needing to learn a new platform.”
Johnny Bytes is an innovative digital agency for web and app development based in Germany. “MySQL HeatWave on AWS simplifies our data platform with a consolidated database for both transaction processing and analytics,” said Thomas Henz, chief executive officer, Johnny Bytes.
“We have seen 60-90X faster complex queries compared to AWS RDS and Aurora that generates real-time analytics we need for targeted, multichannel campaigns. We now have greater scalability to onboard more data and new clients of any size without increasing IT administration.”
As part of the news, Oracle is also introducing several new capabilities and benchmarks for MySQL HeatWave on AWS.
Unmatched performance and price performance: MySQL HeatWave on AWS is optimized for AWS with a superior architecture that delivers higher performance and lower cost compared to competitive offerings, as demonstrated by industry-standard benchmarks. On the 4TB TPC-H benchmark, MySQL HeatWave on AWS delivers price performance that is 7X better than Amazon Redshift, 10X better than Snowflake, 12X better than Google BigQuery, and 4X better than Azure Synapse. For machine learning, MySQL HeatWave on AWS is 25X faster than Redshift ML. On a 10GB TPC-C workload, MySQL HeatWave offers up to 10X higher and sustained throughput compared to Amazon Aurora at high concurrency. All of these fully transparent benchmark scripts are available on GitHub for customers to replicate.
Native AWS experience: MySQL HeatWave on AWS delivers a true native experience for AWS customers through millisecond-level latencies for applications and a rich interactive console. It facilitates schema and data management, and executes queries interactively from the console. Users can monitor the performance of their queries and monitor the utilization of the provisioned resources. MySQL Autopilot is also integrated with the interactive console, making it easier to use.
Advanced security features: MySQL HeatWave service now offers several comprehensive security features which provide additional differentiation with Amazon Aurora. These include server-side data masking and de-identification, asymmetric data encryption, and a database firewall. Asymmetric data encryption enables developers and DBAs to increase the protection of confidential data and implement digital signatures to confirm the identity of people signing documents. Database firewall provides real-time protection against database-specific attacks, such as SQL Injections. These features are designed to provide best in class security for database users and provide a contrast with Aurora, where security methods are layered on top of the database.
MySQL Autopilot: Autopilot provides workload-aware, machine learning-based automation of various aspects of the application lifecycle, including provisioning, data management, query execution, and failure handling. Autopilot features include auto provisioning, auto parallel loading, auto encoding, auto data placement, auto scheduling, auto query plan improvement, auto change propagation, and auto error handling. Combined, these features improve performance of the application, reduce cost by predicting the optimal configuration to run a workload, and reduce manual database administration. Today, Oracle is introducing additional Autopilot capabilities designed for OLTP workloads which further improve MySQL HeatWave price performance compared to Amazon Aurora. Auto thread pooling provides higher and sustained throughput at high concurrency by determining the optimal number of transactions which should be executed. Auto shape prediction determines the optimal shape which should be provisioned to provide the best price performance for OLTP workloads. In a running system, the recommendation could be to continue using the existing shape, to upgrade to a larger shape for better performance or to downgrade to a smaller shape to reduce costs—whichever shape provides the best price performance.
Machine Learning: HeatWave ML provides in-database machine learning capabilities, including training, inference, and explanations. This enables customers to securely use machine learning on real-time data without the complexity, latency, and cost of ETL. HeatWave ML fully automates the ML lifecycle and stores all trained models inside the MySQL database, eliminating the need to move them to a separate machine learning tool or service. It’s available at no additional charge for MySQL HeatWave customers. No other cloud database vendor or open source database provides such advanced ML capabilities inside the database. On average, HeatWave ML trains models 25 times faster than Redshift ML and scales with the cluster size. MySQL HeatWave customers can now train models more often and keep them updated for increased prediction accuracy.