📣 Try out the NEW Gateway to Research and let us know what you think.

We're looking for users to test the new service during August and September and share their feedback. Express your interest by completing this short form.

SACRED-MA: Safe And seCure REmote Direct Memory Access

Lead Research Organisation: IMPERIAL COLLEGE LONDON
Department Name: Computing

Abstract

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

Publications

10 25 50

publication icon
Ambal G (2026) A Verified High-Performance Composable Object Library for Remote Direct Memory Access in Proceedings of the ACM on Programming Languages

publication icon
Ascari F (2026) U-Turn: Enhancing Incorrectness Analysis by Reversing Direction in Proceedings of the ACM on Programming Languages

publication icon
Cho K (2023) Memento: A Framework for Detectable Recoverability in Persistent Memory in Proceedings of the ACM on Programming Languages

publication icon
D'Osualdo E (2023) The Path to Durable Linearizability in Proceedings of the ACM on Programming Languages

publication icon
De Vilhena P (2024) Extending the C/C++ Memory Model with Inline Assembly in Proceedings of the ACM on Programming Languages

 
Description Collaboration with NVIDIA 
Organisation NVIDIA
Country Global 
Sector Private 
PI Contribution A formal semantic model for RDMA programming
Collaborator Contribution Expertise in RDMA programming and understanding of the semantics
Impact We have a preprint that is under submission.
Start Year 2023
 
Description Collaboration with Tel Aviv University 
Organisation Tel Aviv University
Country Israel 
Sector Academic/University 
PI Contribution A formal semantic model for RDMA programming.
Collaborator Contribution Expertise in the area and contacts at NVIDIA
Impact We have a pre-print that is under submission.
Start Year 2023
 
Description University of Colorado, USA 
Organisation University of Colorado Boulder
Country United States 
Sector Academic/University 
PI Contribution I am working with George Hodgkins, Mark Madler and Joe Izraelevitz from University of Colorado to develop a formally verified library for building multi-node objects on RDMA and a modular declarative verification framework called Mowgli with the RDMA memory model, to be used to verify the correctness of LOCO libraries.
Collaborator Contribution They have supported the development of the modular declarative verification framework.
Impact Conference Paper at Principles of Programming Languages (POPL), 2026: A Verified High-Performance Composable Object Library for Remote Direct Memory Access, jointly authored with Guillaume Ambal, George Hodgkins, Mark Madler, Gregory Chockler, Brijesh Dongol, Joe Izraelevitz, and Viktor Vafeiadis https://www.soundandcomplete.org/papers/POPL2026/Mowgli/mowgli.pdf
Start Year 2025
 
Description University of Pisa 
Organisation University of Pisa
Country Italy 
Sector Academic/University 
PI Contribution I am working with Flavio Ascari, Roberto Bruni and Roberta Gori on a combination of Incorrectness Logic and Sufficient Incorrectness Logic to formalize a mixed forward/backward analysis that is able to both highlight the presence of bugs and finding their sources.
Collaborator Contribution We have developed new avenues for embedding incorrectness insights into scalable, expressive, automated code contracts, resulting in our POPL paper U-Turn: Enhancing Incorrectness Analysis by Reversing Direction, published in Proceedings of the ACM on Programming Languages (POPL) 2026
Impact POPL 2026 paper: U-Turn: Enhancing Incorrectness Analysis by Reversing Direction https://dl.acm.org/doi/abs/10.1145/3776688
Start Year 2025
 
Description University of Surrey 
Organisation University of Surrey
Country United Kingdom 
Sector Academic/University 
PI Contribution i am currently collaborating with Brijesh Dongol and Gregory Chockler from University of Surrey on Library of Composable Objects (LOCO), a formally verified library for building multi-node objects on RDMA, which fills the gap between shared memory and distributed system programming. called Mowgli, that is flexible enough to model multinode objects and is independent of a memory consistency model. Work to instantiate Mowgli with the RDMA memory model, and use it to verify correctness of LOCO libraries.
Collaborator Contribution Work on Library of Composable Objects (LOCO), and Mowgli verification framework.
Impact Conference Paper at European Symposium on Programming (ESOP), 2026 (to appear): Specifying and Verifying RDMA Synchronisation, jointly authored with Guillaume Ambal, Max Stupple, Brijesh Dongol https://www.soundandcomplete.org/papers/ESOP2026/rdma-rmw.pdf Conference Paper at Principles of Programming Languages (POPL), 2026: A Verified High-Performance Composable Object Library for Remote Direct Memory Access, jointly authored with Guillaume Ambal, George Hodgkins, Mark Madler, Gregory Chockler, Brijesh Dongol, Joe Izraelevitz, and Viktor Vafeiadis https://www.soundandcomplete.org/papers/POPL2026/Mowgli/mowgli.pdf
Start Year 2024