DTA

Dottorato

PALADINO, FRANCESCO

etd-09252024-135925

Design Optimization and Scheduling of Real-Time Embedded Systems under End-To-End Latency Constraints

ING-INF/05

Istituto di Tecnologie della Comunicazione, dell'Informazione e della Percezione - PHD IN EMERGING DIGITAL TECHNOLOGIES

relatore Prof. BIONDI, ALESSANDRO

• real-time systems
• end-to-end latency
• system design
• optimization
• real-time scheduling

15/03/2025;

parziale

Modern cyber-physical systems offer tons of different functionalities, spanning from careful<br>control activities to infotainment. All these features have different timing properties and<br>requirements: periodicity and deadline may substantially vary in the set of all functionalities,<br>and necessarily guide the design of the application. System designers are required to make<br>some decisions to deploy the application on a hardware platform. First, the features have<br>to be wrapped into OS tasks; then, each task needs to be assigned a priority value to be<br>scheduled by the OS. Moreover, the hardware platform is generally heterogeneous: the<br>available processing units may run at different frequency; besides the CPUs, there may<br>be hardware accelerators like GPUs and FPGAs; the memory banks may be of NUMA<br>type. Placing a functionality on a CPU running at a specific frequency or on a hardware<br>accelerator changes the execution time of that functionality. In addition, depending on the<br>number of tasks running on the same processing unit and their priority, the functionality<br>may suffer from interference. Finally, allocating a variable the functionality uses in a<br>specific memory location affects the access time of that variable. To complicate the picture<br>even more, the functionalities are not isolated, but exchange data. These features can be<br>organized in cause-effect chains where each feature reads data from its producer and writes<br>data to its consumer. Timing requirements can be specified for the end-to-end latency of<br>these chains, i.e., end-to-end deadlines, and finding a way to meet them is a challenging<br>problem. All these design choices cannot be handled manually by system designers; hence,<br>automated design strategies based on optimization techniques are needed. The first part<br>of this thesis proposes a placement method for AUTOSAR classic systems on a NUMA<br>platform with homogeneous processing units. The strategy is based on an MILP formulation<br>and some heuristic algorithms to scale down the complexity of the problem. The second part<br>introduces a design strategy for a more complex scenario, where DAG-based applications<br>with end-to-end deadlines need to be placed on heterogeneous platforms made of CPUs and<br>hardware accelerators like GPUs and FPGAs. An optimal strategy to minimize the energy<br>consumption is proposed based on an MIQCP formulation, while two heuristic algorithms<br>provide suboptimal solutions. The third part describes a mapping strategy for parallel<br>applications developed with the OpenMP library that are modeled as DAGs with end-to-end<br>deadlines. An optimal solution for the deployment problem on platforms with heterogeneous<br>cores is given in the form of an MIQCP formulation that minimizes the power consumption,<br>along with a heuristic algorithm for faster suboptimal configurations. The fourth part of the<br>thesis manages the design of systems where the functionalities are organized in cause-effect<br>chains of periodic LET tasks, and their end-to-end latencies need to be minimized. The<br>proposed strategy assigns priorities to tasks following a branch-and-bound algorithm and,<br>alternatively, heuristic rules for scalability. Finally, the thesis explores the design and<br>development of systems using Lingua Franca, a coordination language that guarantees time<br>determinism and checks at runtime that task deadlines are met. In particular, a layered<br>scheduler was developed to obtain earliest deadline first (EDF) scheduling of reactions in<br>Lingua Franca.

Nome file	Dimensione
Ci sono 1 file riservati su richiesta dell'autore. Contatta l'autore