Project RAPID started off as a hardware-software co-design project targeting large-scale data management and analysis and was targeted to provide the fastest performance for SQL analytics. The project achieved its goals and was demonstrated in a prototype system with a new in-memory analytic engine optimized for custom hardware.
The goal of this project has since been augmented to deliver a system for which provides not just compelling performance for analytic workloads but optimized for performance per dollar. In order to accomplish that, the RAPID software has now been enhanced and optimized for OCI Gen 2 shapes and integrated with the MySQL database.
This technology is being offered as a cloud native MySQL analytic service and provides orders of magnitude improvement in performance over MySQL for analytic queries. Several customers who have tried this service have reported significant improvement in performance and reduction in their cost compared to analytics services offered by other cloud providers.
As a part of this project, we are also exploring many machine learning based techniques for automating various aspects of this service.