Audrey Queudet

Real-Time Support for Software Transactional Memory

Transactional memory is currently a hot research topic, having attracted the focus of both academic researchers and development groups at companies. Indeed, the concept of transactional memory has recently attracted much interest for multicore systems as it eases programming and avoids the problems of lock-based methods. However, up to now, the scheduling of real-time transactions within software transactional memories has not been studied. To address this issue, we present in this paper a real-time software transactional memory, namely RT-STM. We focus on the scheduling of concurrent soft real-time transactions. In particular, we explore a new heuristic for conflict resolution that reduces the number of deadline violations when scheduling soft real-time transactions. After having discussed the scalability of various classical STMs under a real-time operating system, we present experimental results that show that RT-STM can improve the performance of transactional memory-based applications on multicore platforms.

A Note on EDF Schedulingfor Real-Time Energy Harvesting Systems

Energy harvesting is the capture of ambient energy, its conversion into a usable form, and its storage for immediate or future use. Interest in energy harvesting has increased over the last decade because of its environmental friendliness and its ability to power devices without electric wires. This term has been frequently applied in recent years in the context of small autonomous embedded devices such as wireless sensor nodes. In this paper, we address the scheduling problem for a single processor device that executes preemptable time critical tasks. Each one has a certain energy requirement and arrives at an unpredictable time. We ask the question whether the traditional task scheduling algorithm earliest deadline first (EDF) is convenient for energy harvesting environments. The paper shows that EDF has a zero competitive factor but nevertheless is optimal for online non-idling settings.

A Data Flow Monitoring Service Based on Runtime Verification for AUTOSAR

This paper presents the design and implementation of an error detection service for multicore real-time in-vehicle embedded systems. The service aims at monitoring the data flows in a graph of communicating real-time tasks and detecting violation of the expected communication patterns. The service is not based on any specific system model. The monitors are automatically generated from formal models of the monitored system and the expected communication patterns. To minimize the time overhead of the service, the monitors are embedded in the RTOS kernel. The implementation targets an AUTOSAR-like platform based on the open-source RTOS Trampoline. Measures made on an ARM7 MCU show that the time and memory overheads are compatible with the stringent constraints of the application domain.

Clairvoyance and online scheduling in real-time energy harvesting systems

Real-time energy harvesting systems are designed using a microprocessor, a rechargeable energy storage unit and an energy harvester. The theoretical analysis shows that an optimal solution to the underlying online scheduling problem requires time lookahead which can be incompatible with the common stochastic nature of ambient energy.

Chaos-based cryptosystems using dependent diffusion: An overview

Based on some important properties of chaos, such as ergodicity, quasi-randomness, and high sensitivity to the secret key, chaos is a hot research field in secured communication, and recently a variety of chaos-based cryptosystems have been proposed for achieving the confidentiality of transmitted images over public channels. Most of chaos-based encryption algorithms are based on the Fridrich structure, which use separate confusion-diffusion layers. In this paper, we give an overview of some chaos-based block cryptosystems, including our cryptosystem, using dependent diffusion, in which the confusion process and diffusion process are performed sequentially on each pixel of the plain plain image. This kind of cryptosystems are more efficient than the traditional confusion-diffusion architecture of Shannon. A comparison study of efficiency in terms of speed performance and of robustness against cryptanalysis is done.

Task Partitioning Strategies for Multicore Real-Time Energy Harvesting Systems

In this paper, we propose task partitioning heuristics for multicore real-time energy harvesting systems. Both timing constraints and energy requirements are considered. Our objective is to determine a partitioning that guarantees absence of both energy starvation and deadline missing under Earliest Deadline First (EDF) scheduling. First, we describe the system model and formalize the assignment problem in real-time energy harvesting systems. Then, we describe and analyze the performance of partitioning heuristics for identical processors. We explore how the task sorting criteria and the energy constraints can favor the heuristics feasibility performance. Moreover, we estimate through simulation a sufficient energy reservoir capacity that guarantees the best performance.

The BGW model for QoS aware scheduling of real-time embedded systems

Tasks in a real-time computing system are commonly periodic. Each instance generated by the invocation of a periodic task has normally a deadline constraint by which it must complete its execution in all circumstances. However, faults may be present in software, the electrical power supply may be depleted, the processor may be overloaded transiently etc. Any of these emergencies involves a situation where it is no longer possible to meet the deadlines of all the instances in the application. In this paper, we discuss a general model for the application task set that permits the operating system to correctly manage the above emergency situations in accordance with specific parameters statically attached to every task in addition to its classical timing parameters. The model is inspired by the well-known Deadline Mechanism and Skip-Over model. We introduce the BGW model where each instance of any periodic task can get one of the three colours Black, Grey and White. A colour specifies that the instance has to imperatively execute the primary version, the instance has to execute at least one version among primary and alternate or the instance may be discarded. We briefly discuss implementation issues for this new model.

Design and Efficient Implementation of a Chaos-based Stream Cipher

We designed and implemented a stream cipher cryptosystem based on an efficient chaotic generator of finite computing precision (N = 32). The proposed structure of the chaotic generator is formed by a Key-setup, an IVsetup, a non-volatile memory, an output and an internal state function. The cryptographic complexity mainly lies in the internal state containing two recursive filters, with one, two or three delays. Each recursive filter includes a perturbation technique using a linear feedback shift register. The first recursive filter includes a discrete skew tent map, and the second one includes a discrete piecewise linear chaotic map. The chaotic generator is implemented in sequential and parallel versions using Pthread library in a secure manner. The proposed Stream ciphers have very good performance in terms of security and execution time. Experimental results highlight the robustness of the proposed system against known cryptographic and statistical attacks.

Evolution Style: Framework for Dynamic Evolution of Real-Time Software Architecture

Software systems need to be continuously maintained and evolved in order to cope with ever-changing requirements and environments. Introducing these changes without stopping the system is a critical requirement for many software systems. This is especially so when the stop may result in serious damage or monetary losses, hence a mechanism for system change at runtime is needed. With the increase in size and complexity of software systems, software architecture has become the cornerstone in the lifecycle of a software system and constitutes the model that drives the engineering process. Therefore, the evolution of software architecture has been a key issue of software evolution research. Architects have few techniques to help them plan and perform the dynamic evolution of software architecture for real-time systems. Thus, our approach endeavors to capture the essential concepts for modeling dynamic evolution of software architectures, in order to equip the architects with a framework to model this process.

Energy-aware schedulers for real-time energy harvesting systems with quality of service requirements

Our study concerns energy harvesting embedded systems that have real-time constraints. We present two energy aware scheduling algorithms, namely Green-RTO and Green-BWP which aim to optimize the quality of service of the system measured in terms of deadline success ratio. Such algorithms permit to gracefully cope with processing overload and energy starvation. A simulation study permits to show their performance in comparison with the scheduling algorithm EDeg.