ITU researcher secures Carlsberg Foundation grant for query optimisation

Associate Professor Zoi Kaoudi from IT University’s Computer Science Department has secured approximately five million kroner for research in query optimisation.

Computer Science DepartmentawardsZoi Kaoudiartificial intelligence

Written 20 February, 2024 09:11 by Theis Duelund Jensen

Query optimisation is an active research field with a very tangible outcome. Optimising queries in data systems is essential when working towards minimising runtime and the use of processing power which in turn reduces costs and energy consumption.

In recent years, artificial intelligence and machine learning have proven powerful tools for optimisation. That is where associate professor at ITU’s Computer Science Department, Zoi Kaoudi, and her research project Rank4QO: Revolutionizing Query Optimization with Ranking Algorithms come into play.

“Query optimisation is part of any data management system. Basically, whenever you want a computer to do something, execute a command, you are querying the system. In order to be energy efficient and save time, it is in our interest that the system executes queries as efficiently as possible,” says Zoi Kaoudi who has secured a grant of almost five million kroner via Carlsberg Foundation’s Semper Ardens Accelerate programme.

As artificial intelligence and machine learning are becoming more and more effective optimisation tools – reducing manual labour and boosting efficiency – the research project aims to use recent advances to explore the potential of ranking and develop specialised ranking methods to improve query performance.

According to the ITU researcher, we have yet to see the full potential of machine learning and artificial intelligence tools, and so the project will investigate what really matters for efficient query optimisation, i.e., the relative order of query plans and, thus, ranking algorithms. To achieve this, Zoi Kaoudi and her colleagues will conduct a thorough analysis of the connections between queries and query plans, develop learning-to-rank models, and develop new optimisation algorithms.

“Ultimately, our vision is for the project to pioneer a new research paradigm. It is our hope that Rank4QO will reshape the entire query optimization process by incorporating the ranking concept,” says Zoi Kaoudi.

Further information