The Analog Ensemble (AnEn) method can be used to reconstruct incomplete time series using correlated series. Since the AnEn method may use data including several variables through long periods of time, its storage and computational cost may be substantial, slowing down reconstructions. This paper presents a full GPU implementation of the AnEn method, based on PyCUDA, that leads to a significant a cceleration of its execution. The implementation resorts to several techniques that seek to minimize the consumption of GPU global memory in the various steps of the AnEn algorithm, thus making room for larger input datasets. This is further reinforced by the use of batch processing as a way to automatically adapt the datasets size to the GPU memory available. The GPU implementation was tested on a meteorological dataset spanning 10 years, exhibiting a 30-fold speedup in the reconstruction time against a comparable CPUbased multicore version executed with up to 48 cores. The impact on the reconstruction error of changes on several important parameters of the implementation was also assessed.

Flow models over flat porous surfaces have applications in natural processes, such as material, food, chemical processing, or mountain mudflow simulations. The development of simplified analytical or numerical models can predict characteristics such as velocity, pressure, deviation length, and even temperature of such flows for geophysical and engineering purposes. In this context, there is considerable interest in theoretical and experimental models. Mathematical models to represent such phenomena for fluid mechanics have continuously been developed and implemented. Given this, we propose a mathematical and simulation model to describe a free-flowing flow parallel to a porous material and its transition zone. The objective of the application is to analyze the influence of the porous matrix on the flow under different matrix properties. We implement a Computational Fluid Dynamics scheme using the Finite Volume Method to simulate and calculate the numerical solutions for case studies. However, computational applications of this type demand high performance, requiring parallel execution techniques. Due to this, it is necessary to modify the sequential version of the code. So, we propose a methodology describing the steps required to adapt and improve the code. This approach decreases 5.3% the execution time of the sequential version of the code. Next, we adopt OpenMP for parallel versions and instantiate parallel code flows and executions on multi-core. We get a speedup of 10.4 by using 12 threads. The paper provides simulations that offer the correct understanding, modeling, and construction of abrupt transitions between free flow and porous media. The process presented here could expand to the simulations of other porous media problems. Furthermore, customized simulations require little processing time, thanks to parallel processing.

High-Performance Computing involves exploiting the capabilities of powerful computing resources to achieve high computational performance, efficiency, and scalability. By enabling to process complex tasks and/or large amounts of data efficiently, HPC plays an essential role in multiple domains. Parallelization and distribution of workloads across multiple processing resources is at the core of HPC and has been done via many approaches, including different programmatic models and service frameworks. In recent years, HPC has been coupled with Containerization, as a way to deploy, manage and run workloads in an automated and efficient manner, including those relying on accelerators. The performance of heterogeneous workloads, however, depends on the behavior of the platform scheduler, and of the mechanisms used to share accelerators, being also constraint by the limited number of accelerators that each node may host. This paper describes an approach by which the last limitation may be overcome, using a Kubernetes cluster as a show case. The approach builds on the integration of Kubernetes with rOpenCL, an OpenCL API forwarder for remote execution of OpenCL calls. This combination may also be particularly useful in Edge Computing scenarios, where containerized edge services gain the ability to access, transparently and efficiently, remote (e.g. cloud-based) accelerators. Some preliminary experimental results are presented, demonstrating the feasibility of this integration, and the impact on performance for an OpenCL application.

In heterogeneous computing systems, general purpose CPUs are coupled with co-processors of different architectures, like GPUs and FPGAs. Applications may take advantage of this heterogeneous device ensemble to accelerate execution. However, developing heterogeneous applications requires specific programming models, under which applications unfold into code components targeting different computing devices. OpenCL is one of the main programming models for heterogeneous applications, set apart from others due to its openness, vendor independence and support for different co-processors. In the original OpenCL application model, a heterogeneous application starts in a certain host node, and then resorts to the local co-processors attached to that host. Therefore, co-processors at other nodes, networked with the host node, are inaccessible and cannot be used to accelerate the application. rOpenCL (remote OpenCL) overcomes this limitation for a significant set of the OpenCL 1.2 API, offering OpenCL applications transparent access to remote devices through a TPC/IP based network. This paper presents the architecture and the most relevant implementation details of rOpenCL, together with the results of a preliminary set of reference benchmarks. These prove the stability of the current prototype and show that, in many scenarios, the network overhead is smaller than expected.

Abstract Runtime verification is an area of formal methods that studies the dynamic analysis of execution traces against formal specifications. Typically, the two main activities in runtime verification efforts are the process of creating monitors from specifications, and the algorithms for the evaluation of traces against the generated monitors. Other activities involve the instrumentation of the system to generate the trace and the communication between the system under analysis and the monitor. Most of the applications in runtime verification have been focused on the dynamic analysis of software, even though there are many more potential applications to other computational devices and target systems. In this paper we present a collection of challenges for runtime verification extracted from concrete application domains, focusing on the difficulties that must be overcome to tackle these specific challenges. The computational models that characterize these domains require to devise new techniques beyond the current state of the art in runtime verification.

Network inaccessibility is a temporal issue derived from the presence of faults affecting the communication services provided by the medium access control (MAC) sublayer. The occurrence of network inaccessibility represents temporary “communication blackouts”, which prevent communications to be performed and may imply disruptions of network operation, therefore compromising the dependability and timeliness of communications. This paper uses an analytical model accounting for network inaccessibility periods in wireless sensor and actuator networks, presenting the IEEE 802.15.4 standard as a case study. The analytical model is then used to derive a set of simple, yet quite effective policies to reduce the durations of the periods of network inaccessibility. The effectiveness of these policies can be evaluated using a tool based on the analytical model, which is being integrated in the NS-2 simulator for validation. Reducing network inaccessibility is a crucial step to enable the use of wireless networking technologies in real-time settings.

Aerospace mission systems' size, weight and power consumption requirements call for the integration of multiple functions on a single embedded computing platform. A current trend to guard against potential timeliness and safety issues in integrating applications of different natures and providers is the employment of temporal and spatial partitioning. The AIR architecture, defined within initiatives sponsored by the European Space Agency to meet these goals, supports multiple partition operating systems, and advanced timeliness control and adaptation mechanisms. In this paper we propose how to take advantage of composability properties inherent to the build and integration process of AIR-based systems, towards tool-assisted scheduling analysis and configuration support.

The ARINC 653 specification, defined for aeronautical applications, has the goal of providing a standard interface between a given real-time operating system (RTOS) and the corresponding applications. It also provides robust partitioning, with the final goal of guaranteeing safety and timeliness in mission-critical systems. The interest in ARINC 653 has extended to the aerospace industry, which resulted in the definition of an architecture, compliant with the specification, allowing for operating system heterogeneity. In this paper, we introduce the problem of integrating generic operating systems onto this architecture, and explore the case of GNU/Linux. Adding GNU/Linux allows running existing applications or interpreted scripts without needing to port the application or interpreter to an RTOS. In embedded systems, we have to cope with scarce resources and diverse existent hardware, and a balance between both issues must be reached. For such, we show the genesis of such a solution.

Real-time systems usually consist of a set of periodic and sporadic tasks. Periodic tasks can be divided into two classes: synchronous and asynchronous. The first type does not define the task first release, contrary to the second. Hence, synchronous periodic tasks are assumed to be released at the worst instant: the critical instant. The schedulability test is reduced to check a single execution of the task under analysis. The integration of sporadic tasks is also straightforward: they are treated as a periodic task with maximum arrival frequency. On the other hand, asynchronous periodic tasks require a test for each release in the hyper period and the integration of sporadic tasks is not trivial: the worst release instant is unknown a priori. However, they do not assume that the tasks are released at the worst instant.This paper presents a new schedulability analysis method based on the Response Time Analysis (RTA) to determine the worst response time of both asynchronous periodic and sporadic tasks, scheduled by a fixed-priority preemptive algorithm with general deadlines. It also presents another method that enables the introduction of a user configurable degree of pessimism, reducing the hyper period dependency.

The Domus architecture for distributed hash tables (DHTs) is specially designed to support the concurrent deployment of multiple and heterogeneous DHTs, in a dynamic shared-all cluster environment. The execution model is compatible with the simultaneous access of several distributed/parallel client applications to the same or different running DHTs. Support to distributed routing and storage is dynamically configurable per node, as a function of applications requirements, node base resources and the overall cluster communication, memory and storage usage. pDomus is a prototype of Domus that creates an environment where to evaluate the model embedded concepts and planned features. In this paper, we present a series of experiments conduced to obtain figures of merit i) for the performance of basic dictionary operations, and ii) for the storage overhead resulting from several storage technologies. We also formulate a ranking formula that takes into account access patterns of clients to DHTs, to objectively select the most adequate storage technology, as a valuable metric for a wide range of application scenarios. Finally, we also evaluate client applications and services scalability, for a select dictionary operation. Results of the overall evaluation are promising and a motivation for further work

This paper presents VITRAL, a multicolor text mode window manager designed for embedded real-time systems. Therefore, timeliness requirements and resource constraints are the main concern. Even though complex graphical environments are not supported, the application can use VITRAL to build a powerful, dependable environment that reports the system state and still provide a friendly interface to the user.

The CAN Enhanced Layer (CANELy) is a CAN-based infrastructure capable of extremely reliable communication. This paper describes the mechanisms and the techniques used in CANELy to enforce system correctness in the time-domain despite the occurrence of network errors (inaccessibility). The paper discusses how to integrate in the existing CANELy machinery, the control of inaccessibility and the management of timers, at several levels of the system. In particular, application and low-level protocol layers are addressed. In addition, a relevant set of parameters are available for system monitoring, allowing the validation/enforcement of the system model.

Summary form only given. We refine previous work on a model for a distributed hash table (DHT) with support to dynamic balancement across a set of heterogeneous cluster nodes. We present new high-level entities, invariants and algorithms developed to increase the level of parallelism and globally reduce memory utilization. In opposition to a global distribution mechanism, that relies on complete knowledge about the current distribution of the hash table, we adopt a local approach, based on the division of the DHT into separated regions, that possess only partial knowledge of the global hash table. Simulation results confirm the hypothesis that the increasing of parallelism has as counterpart the degradation of the quality of the balancement achieved with the global approach. However, when compared with consistent hashing and our global approach, the same results clarify the relative merits of the extension, showing that, when properly parameterized, the model is still competitive, both in terms of the quality of the distribution and scalability.

The increasing complexity of high-demand long-running applications has faced programmers with the need to take into account both development hardness and execution time, me/spl mu/ provides the flexibility to control the amount of computational and communication power being used in order to maximize resources utilization and to deliver high performance. In this paper we focus on the aspects of the paradigm that go beyond traditional message passing approaches, promoting the idea that by raising the abstraction level of programming models, programmers will make better use of the available resources with clear impact on both productivity and performance. We introduce the resource as the abstraction used to represent and manage both physical resources - nodes, memory, processors and communication technologies - and logical resources - modules, processes, tasks, threads, groups, etc. We also concentrate on the task of specifying, locating and aggregating resources in order to support the mapping of applications into the target cluster hardware and the explicit management of memory hierarchy.

Fault-tolerant distributed systems based on fieldbuses may benefit to a great extent from the availability of semantically rich communication services, such as those provided by group communication, clock synchronization, membership and failure detection. This is specially true of distributed critical control applications. However, the migration of those services to the realm of simple fieldbuses, such as the native Controller Area Network (CAN) protocol family, presents non-negligible problems, since it lacks most of the functionality required of a fault-tolerant distributed system, such as reliable message broadcast guarantees, distributed node failure detection, and site membership services. As part of our endeavor to design a CAN-based infrastructure support for extremely reliable distributed computer control, dubbed CAN Enhanced Layer (CANELy), we have been addressing the problem of fault-tolerant communications on fieldbuses in a comprehensive way. In this paper, we show that node failure detection and site membership services can be efficiently supported by a simple software layer built on top of an exposed CAN controller interface.

Object oriented techniques have been along the last decade one of the most useful programming paradigms. However, for distributed embedded systems, the semantic gap between the object orientation layer and the underlying infrastructure is extremely large. This gap can be narrowed, should the embedded system platform provide semantically rich communication and management services. The paper outlines our research effort in the provision of such services by CAN based (Controller Area Network) systems.

This paper presents a new fault-tolerant clock synchronization algorithm designed for the Controller Area Network (CAN). The algorithm provides all correct processes of the system with a global timebase, despite the occurrence of faults in the network or in a minority of processes. Such global time-frame is a requirement of many distributed real-time control systems. Designing protocols for CAN is justified by the increasing use of this network in industrial automation applications. CAN owns a number of unique properties that can be used to improve the precision and performance of a clock synchronization algorithm. Unfortunately, some of its features also make the implementation of a fault-tolerant clock synchronization service a non-trivial task. Our algorithm addresses both the positive and the negative aspects of CAN.

Presents a processor group membership protocol designed to run on top of a local area network. The protocol maintains information about a selected group of stations that explicitly join the protocol by keeping a replica of a global membership table at every member. Additionally, the protocol guarantees that a given station always occupies the same entry in the table. As a result, table indexes uniquely and universally identify a station and can thus be used as short identifiers. The interest of a processor group membership is twofold: it is a powerful auxiliary for process group membership management and it provides support for efficient message addressing.>

Continuity of service and bounded and known message delivery latency are requirements of a number of applications, which are imperfectly fulfilled by standard LANs. Most previous studies have addressed this issue by computing worst-case access/transmission delays only for normal LAN operation. However, LANs are subject to failures, namely partitions. Since most applications can live with temporary glitches in LAN operation, an alternative approach is to quantify all these glitches or temporary partitions, called inaccessibilities, and to derive a worst-case figure, to be added to the worst-case transmission delay in the absence of faults. In these conditions, reliable real-time operation is possible on nonreplicated LANs. An exhaustive study of the inaccessibility characteristics of the ISO 8802/4 token-bus LAN is described.>