Jade Alglave

2023 United Kingdom Award Finalist — Faculty

Jade Alglave

Current Position:
Professor of Computer Science, University College London; Distinguished Engineer, Arm

Institution:
University College London and Arm

Discipline:
Computer Science

Recognized for: A methodology to develop mathematical models of concurrent systems—systems where multiple calculations are made within overlapping time frames—with the aid of a set of practical software tools, which has significantly influenced computer chips and operating systems.

Areas of Research Interest and Expertise: Concurrency, Semantics, Computer Science

Previous Positions:

BSc and MS, Ecole Normale Superieure de Lyon, France
MS, Master Parisien de Recherche en Informatique, France
PhD, Universite Paris 7 and INRIA, France
Postdoc, University of Oxford
Lecturer, Queen Mary University of London
Researcher, Microsoft Research Cambridge

Research Summary:

Jade Alglave, PhD, has designed a methodology to describe an area of the architecture of modern computers called concurrency. Her work has had a profound impact on the architecture of concurrent systems, such as smart phones, data centers and operating systems.

To increase performance, modern computers feature multiple processors, able to run several jobs concurrently. Those concurrent processors communicate via shared memory. But not all processors see the shared memory in the same way at the same moment: some can already see new, updated data in the memory while others can still only see older data. This makes programming such systems challenging and unintuitive.

Alglave invented a concise and precise mathematical framework that allows to list the general principles and communication rules of concurrent systems. She designed, together with Luc Maranget, PhD, a publicly available and open source set of practical software tools to test whether these specifications are correctly implemented in deployed concurrent systems such as computer chips. Her framework has been applied to many types of concurrent systems, and used by industry leaders, such as Arm, Nvidia, and IBM, and the popular open source operating system Linux. Her framework has helped uncover several bugs in computer chips, both before and after deployment.

"I am thrilled that the methodology and tools that Luc and I have started to develop nearly 15 years ago appear to be useful. I am grateful for UCL giving me the time and space to refine this methodology, and apply it in industry. I am honoured to lead a team at Arm who is doing fantastic work maintaining and developing Arm’s formal concurrency model."

Key Publications: 

  1. Alglave, L. Maranget, M. Tautschnig. Herding Cats: Modelling, Simulation, Testing, and Data-mining for Weak Memory. ACM Transactions on Programming Languages and Systems, 2013.
  2. Alglave, M. Batty, A. Donaldson, G. Gopalakrishnan, J. Ketema, D. Poetzl, T. Sorensen, J. Wickerson. GPU Concurrency: Weak Behaviours and Programming Assumptions. Architectural Support for Programming Languages and Operating Systems, 2015.
  3. Alglave, L. Maranget, P.E. McKenney, A. Parri, A. Stern. Frightening Small Children and Disconcerting Grown-Ups: Concurrency in The Linux Kernel. Architectural Support for Programming Languages and Operating Systems, 2018.
  4. Alglave, W. Deacon, R. Grisenthwaite, A. Hacquard, L. Maranget. Armed Cats: Formal Concurrency Modelling at Arm. ACM Transactions on Programming Languages and Systems, 2021.

Other Honors: 

2021 Fellow, Royal Academy of Engineering
2020 Roger Needham Award, British Computer Society
2018 Silver Medal, Royal Academy of Engineering
2014 Brian Mercer Award for Innovation, The Royal Society

In the Media: 

FENews – The Royal Academy of Engineering Celebrates First New Fellows Elected Under Fit for The Future Diversity Initiative

Website