HeatWave is designed to enable customers to run analytics on data which is stored in MySQL databases without the need for ETL. This service is built on an innovative, in-memory analytics engine which is architected for scalability and performance and is optimized for Oracle Cloud Infrastructure (OCI) Gen 2 hardware. This results in a very performant solution for SQL analytics at a fraction of the cost compared to other cloud services including AWS Aurora, Redshift, Google Big Query, RDS.
The amount of acceleration an application would observe with HeatWave depends upon a number of factors like the datasize, queries, operators being used in the query, the selectivity of the predicates. For the purpose of comparing, we are considering the TPCH benchmark which has the queries well defined and the only variable is the data size and the system configuration. HeatWave is able handle all workloads with a single shape so that significantly simplifies the choice for the customer.
400x Query Acceleration for MySQL The first comparison we make is with MySQL database which is representative of MySQL running on various cloud platforms or various flavors of MySQL. For 400G datasize, using the same number of cores and the same amount of DRAM for MySQL, HeatWave accelerates performance by 400x times for analytics workloads like TPCH. Furthermore, there is no need to create any indexes with HeatWave.
Figure 1. HeatWave accelerates MySQL queries by 400x
1100x Faster than Aurora, 3x cheaper The next comparison we show is with Amazon Aurora, which is Amazon’s premium database service. HeatWave offers dramatic improvement in performance for complex and analytic queries. For a 4TB TPC-H workload, MySQL HeatWave is 1100x faster than Amazon Aurora. Furthermore, there is no need to create indexes on the base table which takes over 5 days with Amazon Aurora compared to under 4 hours to load data in HeatWave. As a result, the data is available to query much sooner than with Aurora. Furthermore, the cost is less than 1/3 of Aurora.
Figure 2. HeatWave is 1100x faster and less than 1/3 the cost of Aurora
The performance improvement of MySQL Database Service with HeatWave over Aurora increases with the size of data.
Figure 3. The performance advantage of HeatWave increases with data size vs. Amazon Aurora
17x Faster than Redshift, 3x Cheaper Next, we compare with Amazon Redshift which is designed for analytics and is offered in multiple shapes. Compared to the fastest shape (dc2.8xlarge), HeatWave is up to 3x faster and 1/3 the cost. For HeatWave, the cost includes both OLTP and OLAP capabilities while for Redshift the additional cost of the OLTP system and the cost of ETL from the OLTP database to Redshift is not included.
Figure 4. HeatWave is 2.7x faster and 1/3 the cost of Amazon Redshift’s fastest shape.
Compared to the cheaper shape of Redshift (RA3.4xLarge), HeatWave is up to 18x faster and 3% less expensive. Unlike Redshift, HeatWave is capable of running both OLTP and OLAP wokloads, without the need for ETL. With Redshift listed cost is only for OLAP, and additional costs are needed for the OLTP database.
Figure 5. HeatWave is 17.7x faster and cheaper than Amazon Redshift’s cheaper shape
Customers who use HeatWave will benefit from significantly better performance, eliminating the need for ETL, support for real-time analytics, reduced monthly cost and a single database for OLTP and OLAP.
Conclusion HeatWave is a cloud native service which is exclusively available in Oracle cloud Infrastructure and provides compelling performance and cost for analytic workloads. Organizations using MySQL database for managing their enterprise data can now run analytic queries with HeatWave with significantly better performance, lower cost, not requiring ETL and support for real- time analytics in contrast to other database services like RDS, Google Big Query, Snowflake, Aurora and Redshift. The service can be deployed in a cloud only or in a hybrid environment, and it simplifies management for both transactional and analytic applications.
We welcome you to try this service for free: https://www.oracle.com/cloud/free/
Learn more about HeatWave: https://www.oracle.com/mysql/heatwave/
Authors
Nipun Agarwal
Senior Vice President, MySQL HeatWave
Nipun Agarwal is Senior Vice President of MySQL Database and HeatWave Development at Oracle. His interests include data processing, machine learning and cloud computing. Prior to this role, Nipun played a pivotal role in Oracle Labs, where he directed various research endeavors that later evolved into new Oracle products, such as MySQL HeatWave. Nipun's journey with Oracle began in 1994, following the completion of his Master of Science in Computer Science. For several years, he contributed to the Oracle database team. To date, Nipun has amassed an impressive portfolio of 200 patent awards.
