I’m extremely grateful to Shopify and the Royal Academy of Engineering for funding my research into Language Engineering! We’re trying a novel approach to solving a real problem around programming language performance that’s encountered by many people in industry and beyond. We know that there will be many research and technical challenges, big and small, to overcome along the way – some we can guess at now, and some will pop up without warning. The crucial thing about the funding from the RAEng and Shopify is that it gives us both the time (five years) and resources (both my time and that of a Research Fellow) to tackle this important problem. I’m particularly looking forward to working with engineers from Shopify, who can help us better understand the effects of our work on large, real world, systems.”
17 May 2022
Royal Academy of Engineering/Shopify Research Chair
Professor Laurence Tratt appointed as Research Chair
Professor Laurence Tratt, Professor of Software Development in the Department of Informatics, was recently awarded a prestigious Research Chair by the Royal Academy of Engineering (RAEng) in conjunction with Shopify, the leading e-commerce provider.
Laurence’s research is about making computer programming language faster, more flexible and cheaper to produce. He is a member of King’s Software Systems Group, which addresses modelling, design and engineering of software systems, automated reasoning about system properties, and mathematical foundations of modelling computing systems.
As RAEng Research Chair Laurence is looking at programming language engineering. He and his team are aiming to improve the performance of programming languages such as Ruby and Python by retrofitting them with state-of-the-art research techniques taken from Just-In-Time compilers (JIT). By ‘meta-tracing’ existing virtual machines (VMs), it is possible to sidestep the problem of manually creating a JIT, while simultaneously guaranteeing 100% compatibility with existing programs. As well as improving programmer productivity this will save energy by reducing the need for servers.
Commenting on the project, Laurence said: