In this paper we present a new algorithm for CPU resource reservation in real-time systems that allows the coexistence of hard, soft and non tasks. The proposed is specifically designed to handle computational overload. A task needs more CPU-time than reserved can reuse spare bandwidth, without interfering with others With respect other reclamation schemes, novelty bandwidth fairly distributed among needing servers. effectiveness demonstrated an extensive set experiments. We also propose...

This paper considers the offline assignment of fixed priorities in partitioned preemptive real-time systems where tasks have precedence constraints. problem is crucial this type systems, as having a good priority allows for an efficient use processing resources while meeting all deadlines. In literature, we can find several proposals to solve problem, which offer varying trade-offs between quality their results and computational complexities. paper, propose new approach, leveraging existing...

Real-time systems are often designed using a set of periodic tasks. Task periods usually by the system requirements, but deadlines and computation times can be modified in order to improve performance. Sensitivity analysis real-time has focused on changes task fixed priority analysis. Only few studies deal with modification dynamic-priority scheduling. The aim this work is provide sensitivity for context dynamic-priority, preemptive, uniprocessor In paper, we present deadline minimization...

In the framework of real-time digital control, two fundamental parameters are defined, control effort and action interval. The first one is related to strength that, due intersampling open-loop determines degrading performances under unexpected delays. second refers unavoidable delays in multitasking environment interactions among tasks. As a consequence, scheduling policy should consider not only tasks but also their influence loop behavior, being calculated minimize overall performances.

Real-time systems are often designed using a set of periodic tasks. Task periods usually by the system requirements, but deadlines and computation times can be modified in order to improve performance. Sensitivity analysis real-time has focused on changes task fixed priority analysis. Only few studies deal with modification dynamic scheduling. The aim this work is provide sensitivity for context dynamic-priority, pre-emptive, uniprocessor In paper, we present deadline minimisation method...

In this paper a new task model with periods defined as ranges is proposed the main goal of drastically reduce hyperperiod set. The focused to be applied in cyclic scheduling, where length major cycle plan determined by hyperperiod. But it also can synthetic sets generation, having small reduces complexity and simulation time. A algorithm, which allows calculate minimum such set tasks, presented. This algorithm calculates value even large number exhaustive search becomes intractable.

One of the most promising approaches to mixed-criticality systems is use multi-core execution platforms based on a hypervisor. Several successful EU Projects are this approach and have overcome some difficulties that introduces. However, interference in COTS due shared resources computer one unsolved problems. In paper, we attempt provide realistic solutions problem. This paper proposes feedback control scheme implemented at hypervisor level transparent partitions (critical non-critical)....

A modern control system is typically implemented as a multitasking software application executing in real-time operating system. If the computer load high, controller will experience delays and jitter, which turn degrade performance. Arguing for an integrated design approach, paper describes two tools implementation-aware analysis: TrueTime Jitterbug. An example given where are used together to evaluate performance of various task implementations.

Abstract In multicore scheduling of hard real-time systems, there is a significant source unpredictability due to the interference caused by sharing hardware resources. This paper deals with schedulability analysis systems where resources taken into account. We rely on task model this integrated in general way, without depending specific type resource. There are similar approaches but they consider fixed priorities. The provided for dynamic priorities assuming constrained deadlines and based...

XtratuM is an open source hypervisor that uses para-virtualization techniques designed to comply with safety critical real-time requirements. Several projects aimed define a reference architecture for space on-board systems have adopted as virtualization layer in order enforce the strong spatial and temporal isolation between software components required on airborne systems. Given shift general trend processor development towards multicore processors, European Space Agency (ESA)...

Embedded systems used in critical systems, such as aeronautics, have undergone continuous evolution recent years. In this evolution, many of the functionalities offered by these been adapted through introduction network services that achieve high levels interconnectivity. The availability access to communications networks has enabled development new applications introduce control functions with higher intelligence and adaptation. applications, it is necessary manage different components an...

Feasibility tests for hard real-time systems provide information about the schedulability of a set tasks. However, this is yes or no answer whether task achieves test not. From system design point view, it would be useful to have more information, example, how much can one vary some parameters, such as computation time, without jeopardizing feasibility. The aim work method determine increase its maintaining feasibility under dynamic priority scheduling. This extra time determined not only in...

Hierarchical scheduling has recently been used to provide temporal isolation embedded virtualised systems. Response time analysis is a common way derive schedulability test for these This paper points out that response hierarchical fixed-priority found in the literature only exact tasks of highest priority domain. For rest an upper bound. In our work, we and compare it with previously published works.

This paper presents a hypervisor architecture tailored to low-power real-time applications. extends the capability of by providing power management techniques and monitoring services. An implementation based on an existing XtratuM that runs over ARM Zynq-7000 SoC device is proposed as proof concept. Measurement results show extended can obtain information consumption reduce it.

In hard real-time embedded systems, switching to multicores is a step that most application domains delay as much possible. This mainly due the number of sources indeterminism, which involve shared hardware resources, such buses, caches, and memories. this paper, new task model considers interference execution causes in other tasks running on cores memory contention proposed. We propose scheduling algorithm calculates exact interference. also analyse compare existing partitioning algorithms...

In the scheduling of hard real-time systems on multicore platforms, significant unpredictability arises from interference caused by shared hardware resources. The objective this paper is to offer a schedulability analysis for such assuming general model that introduces as time parameter each task. assumes constrained deadlines and provided fixed priorities. It based worst-case response analysis, which exists in literature monocore systems. We demonstrate an upper bound, we evaluate our...

En este trabajo se propone la evaluación de diferentes técnicas procesamiento información, obtenida mediante instrumentación, con el objetivo detección anomalías funcionamiento mecánico. Como punto partida usan datos obtenidos muestreo un acelerómetro montado en pórtico XYZ cuyo movimiento está controlado motores paso a paso. A los por instrumentación aplicarán digital, y redes neuronales para determinar posibilidad predicción aparición funcionamientos anómalos máquina. Para experimental las...

Multicore systems introduced a performance increase over previous monocore systems. As these are increasingly finding application in critical domains, it arises necessity to develop novel methodologies for their efficient resource allocation. In addition, is particularly important consider the criticality of applications when scheduling such multicore systems, also includes allocation tasks cores. architectures based on spatial and temporal partitioning, necessary allocate partitions....

Industrial application of digital control requires the synergy between well designed algorithms and carefully implemented systems. The performances can be strongly influenced depending on data acquisition action delays. paper shows how to evaluate these delays (jitter) under static or dynamic scheduling policies. An evaluation several sets tasks executed both policies is analysed compared. results allow us determine goodness with respect due scheduling. While worst case response time in...

The real-time implementation of a controller typically introduces artefacts like delay and jitters that have not been considered at the design stage. As consequence, system behaves in non-periodic manner, real performance is degraded with respect to expected response. This paper proposes hybrid task model reduce impact scheduling on control performance. For large batch typical plants, we analyze how sensitive jitter when sampling rate slow or fast compared bandwidth

Dynamic memory storage has been widely used for years in computer science. However, its use real-time systems not considered as an important issue, and management receive much consideration, whereas today's applications are often characterized by highly fluctuating requirements. In this paper we present approach to dynamic systems. response application behavior requests, the underlying system adjusts resources meet changing demands user needs. The architectural framework that realizes allows...