INRIA

Patrick

Welcome to my home page

Patrick Meumeu Yomsi @ work
Patrick Meumeu Yomsi @ home

INRIA Rocquencourt / AOSTE Team
INRIA

INRIA Rocquencourt
BP 105, Domaine de Voluceau
78153, Le Chesnay Cedex
France

Phone: +33 1 39 63 53 27 / Fax: +33 1 39 63 53 72

Research | Publications | Conferences | Meetings and teaching activities | CV | Others

Research

I am currently a Ph.D. & Expert Engineer at INRIA Paris-Rocquencourt in the AOSTE team. You can find the main goals of our team here. My research activities are in the field of Embedded Systems and concern scheduling approaches in order to accurately take into account the global cost due to preemptions in the analysis of hard real-time systems with precedence, strict periodicity, latency and jitter constraints.

We address the scheduling problem of real-time systems with precedence, strict periodicity, latency and jitter constraints in the monoprocessor case. For such systems, the main challenge for the designer is to basically guarantee that all the deadlines and all constraints are met for all the tasks, otherwise dramatic consequences occur. Guaranteeing deadlines is not always achievable because the preemption is currently approximated within the worst case execution time (WCET) of the tasks when using classical approaches such as RM (Rate Monotonic), DM (Deadline Monotonic), EDF (Earliest Deadline First), LLF (Least Laxity First), etc. This approximation may be wrong since it is difficult to count the exact number of preemptions of each instance for a given task even though the cost of one preemption is easy to know for a given processor. Consequently, this approximation may lead to a wrong real-time execution whereas the schedulability analysis concluded that the system was schedulable. To cope with this problem, the designer usually allows margins which are difficult to assess, and thus in any case lead to a waste of resources. To clearly distinguish between the specification level and its associated model, we use the term operation instead of the commonly used task which is too closely related to the implementation level. Thus, given a set of n periodic preemptive operations with precedence, strict periodicity, latency and jitter constraints, we consider that each operation consists of an infinite sequence of instances, and is characterized by a WCET, not including any approximation of the cost of the preemption, a relative deadline and a period. In the AOSTE team, we are currently seeking for scheduling algorithms which count the exact number of preemptions for each operation, and which lead to new schedulability conditions.

Advisor

I led my research activities under the supervision of Yves Sorel and defended my thesis on April 2nd, 2009 at Université Paris-Sud 11, France.

Publications

Journal papers Conference & Workshop papers Research reports & Thesis Don't look so sad, more are comming up soon :-)
En construction

Submitted papers

Don't look so sad, more are comming up soon :-)

Forthcoming papers

Conferences

Get deadlines of future conferences you might be interested in here.

Meetings and teaching activities

Program Committee Member Teaching activities

Take a look

Last modification : Tuesday 29 August 2006 Send your comments © INRIA Rocquencourt