Browsing by Author "Garibay-Martínez, Ricardo"
Now showing 1 - 6 of 6
Results Per Page
Sort Options
- Architecture to Support Quality of Service in Arrowhead SystemsPublication . Albano, Michele; Garibay-Martínez, Ricardo; Lino Ferreira, LuisThe Arrowhead project [1] considers to normalize all interactions involving embedded systems by mediating them through services. The Service Oriented Architecture (SOA) paradigm is applied to both the interactions that provide the service requested by the user, and other support actions such as the authentication and registration of the devices, and the services they provide, the look-up of devices and service provided, and orchestration of services for creation of more complex services. To this purpose, services are divided into Core Services, which are present in every environment supporting Arrowhead applications, and user services that implement the applications. The Core Services set comprises, at least, Authentication Service, Registration Service and Orchestration Service.
- A framework for the development of parallel and distributed real-time embedded systemsPublication . Garibay-Martínez, Ricardo; Ferreira, Luís; Pinho, Luís MiguelEmbedded real-time applications increasingly present high computation requirements, which need to be completed within specific deadlines, but that present highly variable patterns, depending on the set of data available in a determined instant. The current trend to provide parallel processing in the embedded domain allows providing higher processing power; however, it does not address the variability in the processing pattern. Dimensioning each device for its worst-case scenario implies lower average utilization, and increased available, but unusable, processing in the overall system. A solution for this problem is to extend the parallel execution of the applications, allowing networked nodes to distribute the workload, on peak situations, to neighbour nodes. In this context, this report proposes a framework to develop parallel and distributed real-time embedded applications, transparently using OpenMP and Message Passing Interface (MPI), within a programming model based on OpenMP. The technical report also devises an integrated timing model, which enables the structured reasoning on the timing behaviour of these hybrid architectures.
- Improved Holistic Analysis for Fork–Join Distributed Real-Time Tasks Supported by the FTT-SE ProtocolPublication . Garibay-Martínez, Ricardo; Nelissen, Geoffrey; Lino Ferreira, Luis; Pedreiras, Paulo; Pinho, Luis MiguelModern distributed real-time embedded applications have high processing requirements associated with strict deadlines. For some applications, such constraints cannot be fulfilled by existing single-core embedded platforms. A solution is to parallelize the execution of the applications, by allowing networked nodes to distribute their workload to remote nodes with spare capacity. In that context, this paper presents a holistic timing analysis for fixed-priority fork-join parallel/distributed tasks. Furthermore, we extend the holistic approach to consider the interaction between parallel threads and messages interchanged through a flexible time triggered switched Ethernet network, and we show how the pessimism on the worst case response time computation of such tasks can be reduced by considering the pipeline effect that occurs in such distributed systems. To evaluate the performance and correctness of the holistic model, this paper includes a numerical evaluation based on a real automotive application. The obtained results show that the proposed method is effective in distributing the load by different nodes, allowing a significant reduction of the worst case response time of the tasks. Moreover, the paper also reports an implementation of the model on a Linux library, called parallel/distributed real-time, as well as the corresponding results obtained on a real testbed. The obtained results are in accordance with the predictions of the holistic timing analysis.
- Parallel real-time support for distributed adaptive embedded applicationsPublication . Garibay-Martínez, Ricardo; Ferreira, Luís; Pinho, Luís MiguelReal-time embedded applications require to process large amounts of data within small time windows. Parallelize and distribute workloads adaptively is suitable solution for computational demanding applications. The purpose of the Parallel Real-Time Framework for distributed adaptive embedded systems is to guarantee local and distributed processing of real-time applications. This work identifies some promising research directions for parallel/distributed real-time embedded applications.
- QoS-enabled middleware for smart gridsPublication . Alkhawaja, Abdel Rahman; Ferreira, Luís; Albano, Michele; Garibay-Martínez, RicardoEmerging smart grid systems must be able to react quickly and predictably, adapting their operation to changing energy supply and demand, by controlling energy consuming and energy storage devices. An intrinsic problem with smart grids is that energy produced from in-house renewable sources is affected by fluctuating weather factors. The applications driving smart grids operation must rely on a solid communication network that is secure, highly scalable, and always available. Thus, any communication infrastructure for smart grids should support its potential of producing high quantities of real-time data, with the goal of reacting to state changes by actuating on devices in real-time, while providing Quality of Service (QoS).
- Task partitioning and priority assignment for distributed hard real-time systemsPublication . Garibay-Martínez, Ricardo; Nelissen, Geoffrey; Lino Ferreira, Luís; Pinho, Luís MiguelIn this paper, we propose the Distributed using Optimal Priority Assignment (DOPA) heuristic that finds a feasible partitioning and priority assignment for distributed applications based on the linear transactional model. DOPA partitions the tasks and messages in the distributed system, and makes use of the Optimal Priority Assignment (OPA) algorithm known as Audsley’s algorithm, to find the priorities for that partition. The experimental results show how the use of the OPA algorithm increases in average the number of schedulable tasks and messages in a distributed system when compared to the use of Deadline Monotonic (DM) usually favoured in other works. Afterwards, we extend these results to the assignment of Parallel/Distributed applications and present a second heuristic named Parallel-DOPA (P-DOPA). In that case, we show how the partitioning process can be simplified by using the Distributed Stretch Transformation (DST), a parallel transaction transformation algorithm introduced in [1].