Please contact me for more details.

Postdoc

Implementation of critical applications on multi-core: execution mode analysis to reduce interferences

The implementation of critical applications on multi/many-core platforms is a hot topic in real-time research community. The interferences on shared resources such as shared memory, impacts the execution of critical tasks and their timing analysis. We focus on applications modeled by a direct acyclic graph (DAG) of tasks where edges represent precedence constraints and communications. Our previous work revisits the link between timing analysis and implementation with a collaboration of code orchestration, mapping/scheduling, interference analysis and schedulability. We consider the DAG as a periodic application with one global period. However, applications described by DAGs are generally multi-peridic or subject to execution modes. The idea is to analyze and integrate the execution mode to our implementation/analyses.

The candidate must have knowledge on real-time systems, timing analysis or Synchronous DAG applications. This postdoc is funded by the ANR project CAOTIC

Collaborators: Claire Maiza and Pascal Raymond

Master

Implementation of critical applications on multi-core: execution mode analysis to reduce interferences

Advisers: Claire Maiza and Pascal Raymond

The implementation of critical applications on multi/many-core platforms is a hot topic in real-time research community. The interferences on shared resources such as shared memory, impacts the execution of critical tasks and their timing analysis. We focus on applications modeled by a direct acyclic graph (DAG) of tasks where edges represent precedence constraints and communications. Our previous work revisits the link between timing analysis and implementation with a collaboration of code orchestration, mapping/scheduling, interference analysis and schedulability. We consider the DAG as a periodic application with one global period. However, applications described by DAGs are generally multi-peridic or subject to execution modes. The idea is to analyze and integrate the execution mode to our implementation/analyses. One of these steps will be explored during the Internship.

The candidate must have knowledge on real-time systems, timing analysis or Synchronous DAG applications.

Any older topic below may be discussed also.

Cache analysis

Advisers: David Monniaux, Claire Maïza

In order to prove bounds on the execution time of programs, it is useful to know if memory accesses are in-cache or out of cache. One can do this by static analysis. Obviously, static analysis cannot always conclude (whether an access is in cache may depend on the execution path etc.) but one wants to reduce the cases where one cannot conclude.

For caches with LRU (least recently used) replacement policies, good analysis methods are known. The goal of the internship (which may lead into a PhD) is to look at other policies.

Advanced cache analyzes

Advisers: David Monniaux, Claire Maïza

In order to bound the worst-case execution time (WCET) of programs, or to detect possible side-channel information leaks (security issue), it is necessary to statically analyze their behavior on the memory cache hierarchy.

The topic of the internship is the design and implementation of improved cache analyzes.

Exploration of timing anomaly impact on processor execution for critical real-time systems

details

Advisers: Lionel Rieg, Claire Maiza