Nessy 7.0: entorno de inyección de fallos para Artix-7

El hardware reconfigurable es una tecnología emergente en aplicaciones espaciales.Debido a las características de este hardware, pues su configuración lógica queda almacenada en memoria RAM estática, es susceptible de diversos errores que pueden ocurrir con mayor frecuencia cuando es expuesta a entornos de mayor radiación, como en misiones de exploración espacial. Entre estos se encuentran los llamados SEU o Single Event Upset, y suelen ser generados por partículas cósmicas, pues pueden tener la capacidad de descargar un transistor y de este modo alterar un valor lógico en memoria, y por tanto la configuración lógica del circuito. Por ello que surge la necesidad de desarrollar técnicas que permitan estudiar las vulnerabilidades de diversos circuitos, de forma económica y rápida, además de técnicas de protección de los mismos. En este proyecto nos centraremos en desarrollar una herramienta con este propósito, Nessy 7.0. La plataforma nos permitirá emular, detectar y analizar posibles errores causados por la radiación en los sistemas digitales. Para ello utilizaremos como dispositivo controlador, una Raspberry Pi 3, que contendrá la herramienta principal, y controlará y se comunicará con la FPGA que implementará el diseño a testear, en este caso una placa Nexys 4 DDR con una FPGA Artix-7. Finalmente evaluaremos un par de circuitos con la plataforma.

A Hardware Implementation of a Run-Time Scheduler for Reconfigurable Systems

New generation embedded systems demand high performance, efficiency and flexibility. Reconfigurable hardware can provide all these features. However the costly reconfiguration process and the lack of management support have prevented a broader use of these resources. To solve these issues we have developed a scheduler that deals with task-graphs at run-time, steering its execution in the reconfigurable resources while carrying out both prefetch and replacement techniques that cooperate to hide most of the reconfiguration delays. In our scheduling environment task-graphs are analyzed at design-time to extract useful information. This information is used at run-time to obtain near-optimal schedules, escaping from local-optimum decisions, while only carrying out simple computations. Moreover, we have developed a hardware implementation of the scheduler that applies all the optimization techniques while introducing a delay of only a few clock cycles. In the experiments our scheduler clearly outperforms conventional run-time schedulers based on As-Soon-As-Possible techniques. In addition, our replacement policy, specially designed for reconfigurable systems, achieves almost optimal results both regarding reuse and performance.

A Mapping-Scheduling Algorithm for Hardware Acceleration on Reconfigurable Platforms

Reconfigurable platforms are a promising technology that offers an interesting trade-off between flexibility and performance, which many recent embedded system applications demand, especially in fields such as multimedia processing. These applications typically involve multiple ad-hoc tasks for hardware acceleration, which are usually represented using formalisms such as Data Flow Diagrams (DFDs), Data Flow Graphs (DFGs), Control and Data Flow Graphs (CDFGs) or Petri Nets. However, none of these models is able to capture at the same time the pipeline behavior between tasks (that therefore can coexist in order to minimize the application execution time), their communication patterns, and their data dependencies. This paper proves that the knowledge of all this information can be effectively exploited to reduce the resource requirements and the timing performance of modern reconfigurable systems, where a set of hardware accelerators is used to support the computation. For this purpose, this paper proposes a novel task representation model, named Temporal Constrained Data Flow Diagram (TCDFD), which includes all this information. This paper also presents a mapping-scheduling algorithm that is able to take advantage of the new TCDFD model. It aims at minimizing the dynamic reconfiguration overhead while meeting the communication requirements among the tasks. Experimental results show that the presented approach achieves up to 75% of resources saving and up to 89% of reconfiguration overhead reduction with respect to other state-of-the-art techniques for reconfigurable platforms.

A Hardware Task-Graph Scheduler for Reconfigurable Multi-tasking Systems

Reconfigurable hardware can be used to build a multitasking system where tasks are assigned to HW resources at run-time according to the requirements of the running applications. These tasks are frequently represented as direct acyclic graphs and their execution is typically controlled by an embedded processor that schedules the graph execution. In order to improve the efficiency of the system, the scheduler can apply prefetch and reuse techniques that can greatly reduce the reconfiguration latencies. For an embedded processor all these computations represent a heavy computational load that can significantly reduce the system performance. To overcome this problem we have implemented a HW scheduler using reconfigurable resources. In addition we have implemented both prefetch and replacement techniques that obtain as good results as previous complex SW approaches, while demanding just a few clock cycles to carry out the computations. We consider that the HW cost of the system (in our experiments 3% of a Virtex-II PRO xc2vp30 FPGA) is affordable taking into account the great efficiency of the techniques applied to hide the reconfiguration latency and the negligible run-time penalty introduced by the scheduler computations.

A Replacement Technique to Maximize Task Reuse in Reconfigurable Systems

Dynamically reconfigurable hardware is a promising technology that combines in the same device both the high performance and the flexibility that many recent applications demand. However, one of its main drawbacks is the reconfiguration overhead, which involves important delays in the task execution, usually in the order of hundreds of milliseconds, as well as high energy consumption. One of the most powerful ways to tackle this problem is configuration reuse, since reusing a task does not involve any reconfiguration overhead. In this paper we propose a configuration replacement policy for reconfigurable systems that maximizes task reuse in highly dynamic environments. We have integrated this policy in an external taskgraph execution manager that applies task prefetch by loading and executing the tasks as soon as possible (ASAP). However, we have also modified this ASAP technique in order to make the replacements more flexible, by taking into account the mobility of the tasks and delaying some of the reconfigurations. In addition, this replacement policy is a hybrid design-time/run-time approach, which performs the bulk of the computations at design time in order to save run-time computations. Our results illustrate that the proposed strategy outperforms other state-ofthe-art replacement policies in terms of reuse rates and achieves near-optimal reconfiguration overhead reductions. In addition, by performing the bulk of the computations at design time, we reduce the execution time of the replacement technique by 10 times with respect to an equivalent purely run-time one.

HW implementation of an execution manager for reconfigurable systems

Reconfigurable HW can be used to build a hardware multitasking system where tasks can be assigned to the reconfigurable HW at run-time according to the requirements of the running applications. Normally the execution in this kind of systems is controlled by an embedded processor. In these systems tasks are frequently represented as subtask graphs, where a subtask is the basic scheduling unit that can be assigned to a reconfigurable HW. In order to control the execution of these tasks, the processor must manage at run-time complex data structures, like graphs or linked list, which may generate significant execution-time penalties. In addition, HW/SW communications are frequently a system bottleneck. Hence, it is very interesting to find a way to reduce the run-time SW computations and the HW/SW communications. To this end we have developed a HW execution manager that controls the execution of subtask graphs over a set of reconfigurable units. This manager receives as input a subtask graph coupled to a subtask schedule, and guarantees its proper execution. In addition it includes support to reduce the execution-time overhead due to reconfigurations. With this HW support the execution of task graphs can be managed efficiently generating only very small run-time penalties.

An Approach to Manage Reconfigurations and Reduce Area Cost in Hard Real-Time Reconfigurable Systems

This article presents a methodology to build real-time reconfigurable systems that ensure that all the temporal constraints of a set of applications are met, while optimizing the utilization of the available reconfigurable resources. Starting from a static platform that meets all the real-time deadlines, our approach takes advantage of run-time reconfiguration in order to reduce the area needed while guaranteeing that all the deadlines are still met. This goal is achieved by identifying which tasks must be always ready for execution in order to meet the deadlines, and by means of a methodology that also allows reducing the area requirements.

A Task-Graph Execution Manager for Reconfigurable Multi-tasking Systems

Reconfigurable hardware can be used to build multi tasking systems that dynamically adapt themselves to the requirements of the running applications. This is especially useful in embedded systems, since the available resources are very limited and the reconfigurable hardware can be reused for different applications. In these systems computations are frequently represented as task graphs that are executed taking into account their internal dependencies and the task schedule. The management of the task graph execution is critical for the system performance. In this regard, we have developed two dif erent versions, a software module and a hardware architecture, of a generic task-graph execution manager for reconfigurable multi-tasking systems. The second version reduces the run-time management overheads by almost two orders of magnitude. Hence it is especially suitable for systems with exigent timing constraints. Both versions include specific support to optimize the reconfiguration process.

Reconfigurable computing for Monte Carlo simulations: results and prospects of the Janus project

We describe Janus, a massively parallel FPGA-based computer optimized for the simulation of spin glasses, theoretical models for the behavior of glassy materials. FPGAs (as compared to GPUs or many-core processors) provide a complementary approach to massively parallel computing. In particular, our model problem is formulated in terms of binary variables, and floating-point operations can be (almost) completely avoided. The FPGA architecture allows us to run many independent threads with almost no latencies in memory access, thus updating up to 1024 spins per cycle. We describe Janus in detail and we summarize the physics results obtained in four years of operation of this machine; we discuss two types of physics applications: long simulations on very large systems (which try to mimic and provide understanding about the experimental non equilibrium dynamics), and low-temperature equilibrium simulations using an artificial parallel tempering dynamics. The time scale of our non-equilibrium simulations spans eleven orders of magnitude (from picoseconds to a tenth of a second). On the other hand, our equilibrium simulations are unprecedented both because of the low temperatures reached and for the large systems that we have brought to equilibrium. A finite-time scaling ansatz emerges from the detailed comparison of the two sets of simulations. Janus has made it possible to perform spin glass simulations that would take several decades on more conventional architectures. The paper ends with an assessment of the potential of possible future versions of the Janus architecture, based on state-of-the-art technology.

Interfaz software/hardware entre Raspberry Pi y FPGA Spartan-6 y su aplicación a simulación dirigida por eventos

Hoy día vivimos en la sociedad de la tecnología, en la que la mayoría de las cosas cuentan con uno o varios procesadores y es necesario realizar cómputos para hacer más agradable la vida del ser humano. Esta necesidad nos ha brindado la posibilidad de asistir en la historia a un acontecimiento sin precedentes, en el que la cantidad de transistores era duplicada cada dos años, y con ello, mejorada la velocidad de cómputo (Moore, 1965). Tal acontecimiento nos ha llevado a la situación actual, en la que encontramos placas con la capacidad de los computadores de hace años, consumiendo muchísima menos energía y ocupando muchísimo menos espacio, aunque tales prestaciones quedan un poco escasas para lo que se requiere hoy día. De ahí surge la idea de comunicar placas que se complementan en aspectos en las que ambas se ven limitadas. En nuestro proyecto desarrollaremos una interfaz s oftware/hardware para facilitar la comunicación entre dos placas con distintas prestaciones, a saber, una Raspberry Pi modelo A 2012 y una FPGA Spartan XSA3S1000 con placa extendida XStend Board V3.0. Dicha comunicación se basará en el envío y recepción de bits en serie, y será la Raspberry Pi quien marque las fases de la comunicación. El proyecto se divide en dos partes: La primera parte consiste en el desarrollo de un módulo para el kernel de Linux, que se encarga de gestionar las entradas y salidas de datos de la Raspberry Pi cuando se realizan las pertinentes llamadas de write o read. Mediante el control de los GPIO y la gestión de las distintas señales, se realiza la primera fase de la comunicación. La segunda parte consiste en el desarrollo de un diseño en VHDL para la FPGA, mediante el cual se pueda gestionar la recepción, cómputo y posterior envío de bits, de forma que la Raspberry Pi pueda disponer de los datos una vez hayan sido calculados. Ambas partes han sido desarrolladas bajo licencias libres (GPL) para que estén disponibles a cualquier persona interesada en el desarrollo y que deseen su reutilización.

Herramienta de gestión de contadores hardware para Android

PMCTrack es una herramienta de código abierto para Linux que permite monitorizar el rendimiento de las aplicaciones haciendo uso de los contadores hardware del procesador. Esta herramienta soporta la captura de métricas como el número de instrucciones por ciclo o la tasa de fallos de cache. El objetivo de este proyecto es portar PMCTrack al sistema operativo Android sobre plataformas que integran procesadores de ARM. Esto conlleva la realización de las siguientes tareas: (1) modificación de la variante del kernel Linux propia de Android para incluir las extensiones requeridas por el módulo del kernel de PMCTrack, (2) adaptación de las herramientas de modo usuario de PMCTrack, y (3) desarrollo de una aplicación Android que permita visualizar en tiempo real las medidas de los contadores recabadas para las distintas aplicaciones que están siendo monitorizadas. Para poner a prueba la adaptación de la herramienta PMCTrack al sistema operativo Android y mostrar la utilidad de nuestras aportaciones, se han llevado a cabo diversos casos de estudio empleando la placa de desarrollo Odroid XU4.

Hardware Implementation of a Fault-Tolerant Hopfield Neural Network on FPGAs

This letter presents an FPGA implementation of a fault-tolerant Hopfield NeuralNetwork (HNN). The robustness of this circuit against Single Event Upsets (SEUs) and Single Event Transients (SETs) has been evaluated. Results show the fault tolerance of the proposed design, compared to a previous non fault- tolerant implementation and a solution based on triple modular redundancy (TMR) of a standard HNN design.

A Study of the Robustness Against SEUs of Digital Circuits Implemented with FPGA DSPs

In this paper we present an experimental validation of the reliability increase of digital circuits implemented in XilinxTMFPGAs when they are implemented using the DSPs (Digital Signal Processors) that are available in the reconfigurable device. For this purpose, we have used a fault-injection platform developed by our research group, NESSY [1]. The presented experiments demonstrate that the probability of occurrence of a SEU effect is similar both in the circuits implemented with and without using embedded DSPs. However, the former are more efficient in terms of area usage, which leads to a decrease in the probability of a SEU occurrence.