PhD position on Smart and analyzable scheduling solutions for real-time systems

Apply here.

 

Project description

The project focuses on designing analyzable machine-learning (ML) based scheduling solutions for real-time systems (i.e., systems that require both functional and temporal correctness). What makes this project distinct from the existing learning-based scheduling solutions is that it focuses on the analyzability of the solution so that the result can be used in safety-critical real-time systems, where the end-to-end worst-case response time (WCRT) of each system functionality must be smaller than its deadline to ensure the system’s safety.

To reach this objective, the project focuses on designing scalable and accurate analysis techniques and tools to derive the WCRT of a learning-based scheduler. This can be achieved by defining effective system abstractions that allow performing a scalable yet accurate reachability analysis on the space of all possible system behaviors that could be observed under the proposed smart scheduler.

Our group has experience on designing similar types of analysis for a wide class of scheduling policies such as job-level fixed-priority scheduling (JLFP) policies (we call it “schedule-abstraction based analysis”).

Candidate profiles we are looking for

Required skills. The candidate should have excellent mathematical, computer science and engineering skills and have affinity with scheduling and real-time systems. Experience on formal methods (or knowledge about system verification) is definitely a plus.

The candidate should be highly motivated and eager to learn new topics and be able to acquire the knowledge she/he needs very fast to be able to achieve good results in this project.

This project targets a challenging yet very hot topic in safety-critical systems and hence provides ample opportunities for the candidate to accomplish impactful scientific results that are highly visible. If you like to take this challenge and you think you have the right skills, then you may just be the candidate that we are looking for.

Application

Apply through the TU/e website (vacancy number 914870):

·        A cover letter explaining your motivation and suitability for the position;

·        A detailed Curriculum Vitae (including a list of publications and key achievements);

·        A written scientific report in English of which you are the main author (MSc thesis, traineeship report or scientific paper);

·        Contact information of two references;

·        Copies of diplomas with course grades (transcripts).

What is a schedule-abstraction graph and how does it work?

Schedule-abstraction graph (SAG) explores the space of possible decisions that a job-level fixed-priority (JLFP) scheduler can take when dispatching a set of jobs on processing resources. This decision space is explored by building a graph whose vertices represent the state of the resource (e.g., processor) after the execution of a set of jobs. The edges of this graph represent possible scheduling decisions that evolve the system states.

SAG has been designed for non-preemptive jobs [Nasri2017, Nasri2018, Nasri2019, Nogd2020, Ranjha2021, Ranjha2022], hence, a scheduling decision is to determine a next job that can possibly be dispatched' after a system state.

Want to learn more?

·        An introduction to schedule-abstraction graph analysis: slides

·        An introduction on my research: slides

Implementation

Schedule-abstraction graph is an open-source analysis tool and is available on github. The repository is maintained by Dr. Geoffrey Nelissen.

 

Collaborators

Current student team

Publications

My publications in Google Scholar.

Technical reports

Peer Reviewed Conference Papers

Theses