The Panasonic Group has launched Panasonic Transformation (PX) as a group-wide initiative centered on DX (Digital Transformation). In addition to IT modernization, the group is promoting the creation of an open and flat organizational culture that emphasizes diversity, fairness, and inclusivity, as well as transforming business processes.
To leverage data, a crucial management resource for our group, as a business asset, we are implementing various IT transformation measures, including system modernization, cloud adoption, the construction of a data-driven platform, and supply chain optimization. In the future, we aim to accumulate expertise internally and provide value to external clients in transforming their businesses.
The Panasonic Group operates nearly 2,000 database environments of varying sizes on a daily basis. Since 2011, we have been operating more than 25 on-premises Oracle Exadata databases, and by 2020, we were running more than 60 databases on AWS RDS, demonstrating our commitment to cloud utilization. From 2023, we are gradually migrating our core systems to Exadata Database Service on Oracle Cloud, accelerating data utilization in a multi-cloud environment.
We had several systems built with MySQL within our company, and we were considering how to migrate them to the cloud. We were also considering three data warehouse (DWH) products other than Exadata as an analysis platform for newly acquired data, and one of them caught our attention: HeatWave, which integrates transactional/analytics/machine learning/generative AI in a single engine. We thought that being able to complete data integration, analysis, and utilization on a single platform would be an advantage for using data in business. The Panasonic Group uses AWS and Oracle Cloud, and compatibility with both environments was also a deciding factor. We had been regularly gathering information on product benefits and other things from HeatWave’s user community, “HeatWavejp,” so we decided to proceed with a full-scale evaluation as a candidate for our analysis platform.
We had previously interacted with Smart Style through technical exchanges at external events, and they had also supported us with our MySQL system in the past. We had a high level of trust in them due to their technical expertise and extensive experience in the MySQL field. This time, they supported us from building the PoC environment to organizing information for the evaluation report, allowing us to focus on considering the business requirements.
Although some members of the testing team were not familiar with MySQL, Smart Style provided thorough instruction on the basic functions of MySQL and HeatWave, which allowed us to proceed with the project smoothly.
In addition to serving as the foundation for the group’s critical systems, we also conducted testing with the intention of using it as an external service platform. Specifically, the key points of the testing were whether it could meet internal operational standards and whether availability and security could be guaranteed.
In terms of operation, some necessary functions for performance monitoring and audit log acquisition were lacking, and we had to create our own scripts to supplement these functions to meet our operational standards. SmartStyle has forwarded this request to the manufacturer, and we hope that these functions will be added in the future.
In terms of security, HeatWave has the same level of security features as MySQL Enterprise Edition, so we believe it can be applied to high-priority systems without any problems.
HeatWave’s Autopilot Indexing feature is appealing because it uses machine learning to suggest the optimal index, significantly reducing the effort required for tuning while improving performance. Testing slow queries with this feature resulted in a 50% reduction in processing time. GenAI, which can respond to commands and workflows in natural language after being trained on procedure manuals and operational guides, has the potential to significantly reduce the effort required for handling inquiries.
