The School of Computing is delighted to offer a new PhD studentship entitled: ‘Catching and Mitigating Event-Loop Concurrency Issues‘.
Modern server applications utilize concurrency in many ways. Actors and event loops are popular as programming paradigms since they avoid low-level concurrency issues. Unfortunately, even when using actors or event loops, server applications still show other types of concurrency bugs. Though, they can be mitigated with run-time detection techniques. In the CaMELot project, we want to make detection and mitigation fast enough to be useable in production systems. This way, even though software may contain bugs, we can reduce their impact on correct execution.
Building on a system for deterministic replay of actor applications, you would investigate how one can use lightweight race detection to facilitate debugging of the underlying concurrency issues. Afterall, if we can detect the issues, we should help developers to fix them! One of the challenges here is to derive actionable insights from the detected races. A tool processing the race detection results may guide developers to fix the underlying issues, possibly replaying the problematic execution for the developers.
The candidate will be based in the School of Computing and will work under the supervision of Dr Stefan Marr. Funding is available at the current home fee rate of £4,407.00 together with a maintenance grant of £15,285.00, and is awarded as a Graduate Teaching Assistantship (GTA). The successful candidate will be required to start in September 2021.
The application deadline is Wednesday 13 January 2021. Enquiries relating to the project should be directed to Dr Stefan Marr at firstname.lastname@example.org.
Further details about the studentship, including the application link, can be found here: