J. Javier Gutiérrez

Schedulability Analysis and Optimization of Heterogeneous EDF and FP Distributed Real-Time Systems

The increasing acceptance of the Earliest Deadline First (EDF) scheduling algorithm in industrial environments, together with the continued usage of Fixed Priority (FP) scheduling is leading to heterogeneous systems with different scheduling policies in the same distributed system. Schedulability analysis techniques usually consider the entire system as a whole (holistic approach), with only one preestablished scheduling policy in all the resources. In this work, composition mechanisms will be presented that enable us to combine different FP and EDF response-time analysis techniques for checking the schedulability of heterogeneous systems. Additionally, priority and scheduling deadline assignment techniques will be combined into a new algorithm called HOSPA (Heuristic Optimized Scheduling Parameters Assignment), for optimizing the assignment of priorities and scheduling deadlines to tasks and messages in heterogeneous distributed hard real-time systems.

Modeling distributed real-time systems with MAST 2

Abstract Switched networks have an increasingly important role in real-time communications. The IEEE Ethernet standards have defined prioritized traffic (802.1p) and other QoS mechanisms (802.1q). The Avionics Full-Duplex Switched Ethernet (AFDX) standard defines a hard real-time network based on switched Ethernet. Clock synchronization is also an important service in some real-time distributed systems because it allows a global notion of time for event timing and timing requirements. In the process of defining the new MAST 2 model, clock synchronization modeling capabilities have been added, and the network elements have been enhanced to include switches and routers. This paper introduces the schedulability model that will enable an automatic schedulability analysis of a distributed application using switched networks and clock synchronization mechanisms.

Holistic schedulability analysis for multipacket messages in AFDX networks

The ARINC-664, Part 7 (AFDX) standard defines a communication network based on Ethernet and the UDP/IP protocols. Contrary to general-purpose Ethernet, the timing behavior in AFDX is deterministic due to the use of special network switches and end systems with static routing tables and traffic policing at the sending end through mechanisms called virtual links. Even though the latencies in this network are bounded, there are scheduling and contention effects that need to be analyzed. In this paper we develop a response-time analysis for multipacket messages transmitted through an AFDX network including the scheduling of the virtual links and sub-virtual links, and also the contention in the end systems and in the switches. This analysis allows us to obtain worst-case latencies and output jitter for the network messages with a precise modeling of the sending and receiving ends. These results can be integrated in a holistic approach with the response time analysis of the threads in the processing nodes to obtain end-to-end response times in heterogeneous distributed systems.

A survey on standards for real-time distribution middleware

This survey covers distribution standards oriented to the development of distributed real-time systems. Currently, there are many distribution standards that provide a wide and different set of real-time facilities to control the temporal aspects of applications. Besides giving a general overview of these standards, we describe the real-time mechanisms proposed by each standard to manage both processor and network resources, discuss whether the available facilities are sufficient to guarantee determinism throughout the whole application, and identify a set of features and deployment options that would be desirable in any real-time distribution middleware regardless of its distribution model and standard. The survey identifies open issues and key challenges for future research.

The Chance for Ada to Support Distribution and Real-Time in Embedded Systems

This paper presents a modification of GLADE – the current GNAT implementation of the Ada 95 Distributed Systems Annex (DSA) – to support the development of distributed applications with hard real-time requirements. This modified implementation, that we call RT-GLADE (Real-Time GLADE), is specially suitable for embedded applications composed of a small number of heterogeneous processors and communication networks, because it ensures pre-dictable timing behaviour. A real-time model of the implementation allows the application developer to determine and optimize the overall timing behaviour by applying the corresponding schedulability analysis and priority assignment tech-niques. This realtime version of GLADE continues to conform to the DSA, so the entire real-time application can be built within the Ada 95 context. To imple-ment RT-GLADE, we provide a priority-based communication network over standard Ethernet that is used to ensure predictable transmission times.

Permutational Genetic Algorithm for the Optimized Assignment of Priorities to Tasks and Messages in Distributed Real-Time Systems

The assignment of fixed priorities to tasks and messages in distributed real-time systems is known to be an NP-hard problem, and thus there is no optimal method to accomplish it in polynomial time. This fact makes it a suitable problem to be approached with generic search and optimization algorithms. In this paper we propose a genetic algorithm with a per mutational solution encoding for the assignment of fixed priorities to tasks and messages in distributed real-time systems using a holistic approach. This paper shows that the genetic algorithm can find more and better schedulable priority assignments than HOPA, which is, as far as we know, one of the best methods for the fixed priority assignment in distributed real-time systems.

Interchangeable scheduling policies in real-time middleware for distribution

When a middleware layer is designed for providing semi-transparent distribution facilities to real-time applications, a trade-off must be made between the expressiveness and control capabilities of the real-time parameters used, and the simplicity of usage. Middleware specifications such as RT-CORBA or Adas Distributed Systems Annex (DSA) rely on the use of priorities to map the timing requirements of the application, thus restricting the possible scheduling policies. This paper presents a generic technique to express complex scheduling and timing parameters of distributed transactions, allowing real-time middleware implementations to change their scheduling policies for both the processing nodes and the networks. The technique has been tested in an implementation of Adas DSA, providing two interchangeable policies: a fixed-priority scheduler, and a complex contract-based flexible scheduler.

Experience in integrating interchangeable scheduling policies into a distribution middleware for Ada

This paper presents the integration of real-time scheduling mechanisms within middleware which supports distribution through Ada and CORBA standards. This technology includes several aspects closely related to real-time systems, such as the management of remote calls, identification of schedulable entities, support for the transactional model and the assignment of scheduling parameters. As a result, a real-time distribution middleware capable of managing different scheduling policies under a generic interface has been obtained. The paper also presents a case study in order to evaluate the proposed technology.1

Real-Time Distribution Middleware from the Ada Perspective

Standards for distribution middleware sometimes impose restrictions and often allow the implementations to decide on aspects that are fundamental to the correct and efficient behaviour of the applications using them, especially when these applications have real-time requirements. This work presents a study of two standard approaches for distribution middleware that can be used from Ada applications: RT-CORBA, and the Distributed Systems Annex (DSA) of Ada. The study focuses on the problems associated with the real-time behaviour of some implementations of these approaches, and on possible solutions that can be derived from our experience with Ada implementations. Moreover, the paper considers the problem of integration of the distribution middleware with a new generation of scheduling mechanisms based on contracts.

Deadline Assignment in EDF Schedulers for Real-Time Distributed Systems

Real-time distributed systems contain end-to-end flows, which are distributed actions composed of sequences of tasks activated through messages. Such flows usually have an end-to-end deadline but the internal tasks and messages do not have specific timing requirements. However, if EDF schedulers are used, it is necessary to assign scheduling deadlines to tasks and messages, which is usually done by distributing the end-to-end deadline among them. Distributed systems may have synchronized global clocks or non-synchronized local clocks. This work studies the influence of the clocks, global or local, on the deadline-assignment algorithms. A study on the poor performance observed for EDF schedulers with local clocks is presented. Then, a significant optimization of the assignment algorithms is shown, in which an amount of end-to-end deadline larger than the established timing requirement is distributed among tasks and messages. With this technique, two new algorithms for deadline-assignment are proposed, showing that they outperform the existing ones by up to 23 percent of processor utilization in the case of local clocks. Finally, the influence of release jitter in this kind of EDF systems and the positive effects of eliminating it are also studied.