Scientific objectives
Objective 1: worst case execution time estimation of a program
Modern processors induce an increased variability of the execution time of programs, making difficult (or even impossible) a complete static analysis. Our objective is to propose a solution composing probabilistic and non-probabilistic approaches based on static analyses locally and on statistical analyses globally by answering the following identified scientific challenges:
- a classification of the variability of execution times of a program with respect to the processor features. The access to this later page requires a login (aoste) and a password (aoste). The difficulty of this challenge is related to the definition of an element belonging to the set of variability factors and its mapping to the execution time of the program.
- a compositional rule of statistical models based on Bayesian approaches. The difficulty of this challenge comes from the fact that a global maximum cannot be obtained by upperbouding the corresponding local maxima.
Objective 2: deciding the schedulability of all programs running within the same cyber component
The programs may have different time criticalities, but they share the same processor, possibly multicore. This later case is referred as a mixed criticality approach. Our objective is to propose a solution composing probabilistic and non-probabilistic approaches based on answers to the following identified scientific challenges:
- scheduling algorithms taking into account the interaction between different variability factors. The existence of time parameters described by probability distributions imposes to answer to the challenge of revisiting scheduling algorithms that lose their optimality even in the case of a single core processor. Moreover, the multicore partionning problem is recognized difficult for the non-probabilistic case.
- schedulability analyses based on the algorithms proposed previously. In the case of predictable processors, the schedulability analyses accounting for operating systems costs increase the dependability of CPSs. Moreover, in presence of variability factors, the composition property of non-probabilistic approaches is lost and new principles are required.
Objective 3: deciding the schedulability of all programs communicating through predictable and non-predictable networks
In this case the programs of the same cyber component execute on the same processor and they may communicate with the programs of other cyber components through networks that may be predictable (network on chip) or non-predictable (internet, telecommunications). Our solution is built by analysing schedulability of programs, for which existing (worst case) probabilistic solutions exist, communicating through networks, for which probabilistic worst-case solutions and average solutions exist.
Objective 4: minimizing the energy consumption
Intuitively the statistical approaches may optimize the CPSs design and the energy consumption is one possible way to quantify the expected gain. The difficulty in achieving such objective comes from the fact that all current models are including CPU frequency variation, when the largest energy consumption feature is the memory access. Our solution is built by proposing energy-oriented models.
