DBtune: Database tuning software as a service
Luigi Nardi is an assistant professor of machine learning at Lund University and a research staff at Stanford University. His current research focuses on black-box optimization theory and its use in various practical applications, including AutoML, hardware design, database management, computer vision, and robotics. Prior to joining the faculty at LU, Luigi was a post-doc in the department of computing at Imperial College London and worked as a software engineer at the financial firm Murex S.A.S. after completing his Ph.D. in 2011 at Université Pierre et Marie Curie (UPMC) in Paris. Luigi is the founder and CEO of DBtune (www.dbtune.ai).
No video of the event yet, sorry!
Database management systems expose configurable parameters that control their runtime behavior. The general trend is that over the years, new parameters are added in each new system release, resulting in an explosion of parameters, e.g., PostgreSQL has hundreds of free parameters to date. Setting these parameters correctly improves the performance of an application and saves substantial cloud resources. Given the complex interaction of these parameters, tuning them requires considerable effort from experienced database administrators. That’s where DBtune can help. DBtune is a fully automated service enabling database parameter tuning. The optimizer is based on machine learning technology that customizes each optimization to the specific customer workload and hardware in use. DBtune observes the KPIs of a database and trains recommendation models used to select parameters that lead to better KPIs, making the process of database tuning both more efficient and scalable to thousands of instances. Using DBtune, users can dramatically reduce the cloud costs and the CO2 impact of their database instances.
- 2023 April 20 16:50 PDT
- 20 min
- San Pedro, Lvl C
- Silicon Valley 2023