The VINEYARD Approach: Versatile, Integrated, Accelerator-Based, Heterogeneous Data Centres.

Emerging web applications like cloud computing, Big Data and social networks have created the need for powerful centres hosting hundreds of thousands of servers. Currently, the data centres are based on general purpose processors that provide high flexibility buts lack the energy efficiency of customized accelerators. VINEYARD aims to develop an integrated platform for energy-efficient data centres based on new servers with novel, coarse-grain and fine-grain, programmable hardware accelerators. It will, also, build a high-level programming framework for allowing end-users to seamlessly utilize these accelerators in heterogeneous computing systems by employing typical data-centre programming frameworks (e.g. MapReduce, Storm, Spark, etc.). This programming framework will, further, allow the hardware accelerators to be swapped in and out of the heterogeneous infrastructure so as to offer high flexibility and energy efficiency. VINEYARD will foster the expansion of the soft-IP core industry, currently limited in the embedded systems, to the data-centre market. VINEYARD plans to demonstrate the advantages of its approach in three real use-cases (a) a bio-informatics application for high-accuracy brain modeling, (b) two critical financial applications, and (c) a big-data analysis application.

Quantitative evaluation of three reconfiguration strategies on FPGAs: a case study

Reconfigurable computing is becoming an important new alternative for implementing computations. Field programmable gate arrays (FPGAs) are the ideal integrated circuit technology to experiment with the potential benefits of using different strategies of circuit specialization by reconfiguration. The final form of the reconfiguration strategy is often non-trivial to determine. Consequently, in this paper, we examine strategies for reconfiguration and, based on our experience, propose general guidelines for the tradeoffs using an area-time metric called functional density. Three experiments are set up to explore different reconfiguration strategies for FPGAs applied to a systolic implementation of a scalar quantizer used as a case study. Quantitative results for each experiment are given. The regular nature of the example means that the results can be generalized to a wide class of industry-relevant problems based on arrays.

A self-healing real-time system based on run-time self-reconfiguration

The new generations of SRAM-based FPGA (field programmable gate array) devices are the preferred choice for the implementation of reconfigurable computing platforms intended to accelerate processing in real-time systems. However, FPGA's vulnerability to hard and soft errors is a major weakness to robust configurable system design. In this paper, a novel built-in self-healing (BISH) methodology, based on run-time self-reconfiguration, is proposed. A soft microprocessor core implemented in the FPGA is responsible for the management and execution of all the BISH procedures. Fault detection and diagnosis is followed by repairing actions, taking advantage of the dynamic reconfiguration features offered by new FPGA families. Meanwhile, modular redundancy assures that the system still works correctly

JPEG decoder implementation on FPGA using dynamic partial reconfiguration

Trabalho Final de Mestrado para obtenção do grau de Mestre em Engenharia de Electrónica e telecomunicações

A clocking technique with power savings in virtex-based pipelined designs

This paper presents the evaluation in power consumption of a clocking technique for pipelined designs. The technique shows a dynamic power consumption saving of around 30% over a conventional global clocking mechanism. The results were obtained from a series of experiments of a systolic circuit implemented in Virtex-II devices. The conversion from a global-clocked pipelined design to the proposed technique is straightforward, preserving the original datapath design. The savings can be used immediately either as a power reduction benefit or to increase the frequency of operation of a design for the same power consumption.

Quantitative evaluation of three reconfiguration strategies on FPGAs: a case study

Reconfigurable computing is becoming an important new alternative for implementing computations. Field programmable gate arrays (FPGAs) are the ideal integrated circuit technology to experiment with the potential benefits of using different strategies of circuit specialization by reconfiguration. The final form of the reconfiguration strategy is often non-trivial to determine. Consequently, in this paper, we examine strategies for reconfiguration and, based on our experience, propose general guidelines for the tradeoffs using an area-time metric called functional density. Three experiments are set up to explore different reconfiguration strategies for FPGAs applied to a systolic implementation of a scalar quantizer used as a case study. Quantitative results for each experiment are given. The regular nature of the example means that the results can be generalized to a wide class of industry-relevant problems based on arrays.

Contribuição para o estudo do embarque de uma rede neural artificial em field programmable gate array (FPGA)

This study shows the implementation and the embedding of an Artificial Neural Network (ANN) in hardware, or in a programmable device, as a field programmable gate array (FPGA). This work allowed the exploration of different implementations, described in VHDL, of multilayer perceptrons ANN. Due to the parallelism inherent to ANNs, there are disadvantages in software implementations due to the sequential nature of the Von Neumann architectures. As an alternative to this problem, there is a hardware implementation that allows to exploit all the parallelism implicit in this model. Currently, there is an increase in use of FPGAs as a platform to implement neural networks in hardware, exploiting the high processing power, low cost, ease of programming and ability to reconfigure the circuit, allowing the network to adapt to different applications. Given this context, the aim is to develop arrays of neural networks in hardware, a flexible architecture, in which it is possible to add or remove neurons, and mainly, modify the network topology, in order to enable a modular network of fixed-point arithmetic in a FPGA. Five synthesis of VHDL descriptions were produced: two for the neuron with one or two entrances, and three different architectures of ANN. The descriptions of the used architectures became very modular, easily allowing the increase or decrease of the number of neurons. As a result, some complete neural networks were implemented in FPGA, in fixed-point arithmetic, with a high-capacity parallel processing

Towards a java bytecodes compiler for nios II soft-core processor

Reconfigurable computing is one of the most recent research topics in computer science. The Altera - Nios II soft-core processor can be included in a large set of reconfigurable architectures, especially because it is designed in software, allowing it to be configured according to the application. The recent growth in applications that demand reconfigurable computing made necessary the building of compilers that translate high level languages source codes into reconfigurable devices instruction sets. In this paper we present a compiler that takes as input the bytecodes generated by a Java front-end compiler and generates a set of instructions that attends to the Nios II processor instruction set rules. Our work shows how we process Java bytecodes to the intermediate code, in the Nios II instructions format, and build the control flow and the control dependence graphs. © 2009 IEEE.

Compilação de bytecodes Java para um ambiente de arquitetura reconfigurável

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

LALP: a language to program custom FPGA-based acceleration engines

Field-Programmable Gate Arrays (FPGAs) are becoming increasingly important in embedded and high-performance computing systems. They allow performance levels close to the ones obtained with Application-Specific Integrated Circuits, while still keeping design and implementation flexibility. However, to efficiently program FPGAs, one needs the expertise of hardware developers in order to master hardware description languages (HDLs) such as VHDL or Verilog. Attempts to furnish a high-level compilation flow (e.g., from C programs) still have to address open issues before broader efficient results can be obtained. Bearing in mind an FPGA available resources, it has been developed LALP (Language for Aggressive Loop Pipelining), a novel language to program FPGA-based accelerators, and its compilation framework, including mapping capabilities. The main ideas behind LALP are to provide a higher abstraction level than HDLs, to exploit the intrinsic parallelism of hardware resources, and to allow the programmer to control execution stages whenever the compiler techniques are unable to generate efficient implementations. Those features are particularly useful to implement loop pipelining, a well regarded technique used to accelerate computations in several application domains. This paper describes LALP, and shows how it can be used to achieve high-performance computing solutions.

Multi Processor Systems On Chip with Configurable Hardware Acceleration

The evolution of the electronics embedded applications forces electronics systems designers to match their ever increasing requirements. This evolution pushes the computational power of digital signal processing systems, as well as the energy required to accomplish the computations, due to the increasing mobility of such applications. Current approaches used to match these requirements relies on the adoption of application specific signal processors. Such kind of devices exploits powerful accelerators, which are able to match both performance and energy requirements. On the other hand, the too high specificity of such accelerators often results in a lack of flexibility which affects non-recurrent engineering costs, time to market, and market volumes too. The state of the art mainly proposes two solutions to overcome these issues with the ambition of delivering reasonable performance and energy efficiency: reconfigurable computing and multi-processors computing. All of these solutions benefits from the post-fabrication programmability, that definitively results in an increased flexibility. Nevertheless, the gap between these approaches and dedicated hardware is still too high for many application domains, especially when targeting the mobile world. In this scenario, flexible and energy efficient acceleration can be achieved by merging these two computational paradigms, in order to address all the above introduced constraints. This thesis focuses on the exploration of the design and application spectrum of reconfigurable computing, exploited as application specific accelerators for multi-processors systems on chip. More specifically, it introduces a reconfigurable digital signal processor featuring a heterogeneous set of reconfigurable engines, and a homogeneous multi-core system, exploiting three different flavours of reconfigurable and mask-programmable technologies as implementation platform for applications specific accelerators. In this work, the various trade-offs concerning the utilization multi-core platforms and the different configuration technologies are explored, characterizing the design space of the proposed approach in terms of programmability, performance, energy efficiency and manufacturing costs.

A Precharge-Absorbed DPL Logic for Reducing Early Propagation Effects on FPGA Implementations

In this paper, a new countermeasure against power and electromagnetic (EM) Side Channel Attacks (SCA) on FPGA implemented cryptographic algorithms is proposed. This structure mainly focuses on a critical vulnerability, Early Evaluation, also known as Early Propagation Effect (EPE), which exists in most conventional SCA-hardened DPL (Dual-rail with Precharge Logic) solutions. The main merit of this proposal is that the EPE can be effectively prevented by using a synchronized non regular precharge network, which maintains identical routing between the original and mirror parts, where costs and design complexity compared with previous EPE-resistant countermeasures are reduced, while security level is not sacrificed. Another advantage for our Precharge Absorbed(PA) - DPL method is that its Dual-Core style (independent architecture for true and false parts) could be generated using partial reconfiguration. This helps to get a dynamic security protection with better energy planning. That means system only keeps the true part which fulfills the normal en/decryption task in low security level, and reconfigures the false parts once high security level is required. A relatively limited clock speed is a compromise, since signal propagation is restricted to a portion of the clock period. In this paper, we explain the principles of PA-DPL and provide the guidelines to design this structure. We experimentally validate our methods in a minimized AES co-processor on Xilinx Virtex-5 board using electromagnetic (EM) attacks.

Fault Tolerance Analysis and Self-Healing Strategy of Autonomous, Evolvable Hardware Systems

This paper presents an analysis of the fault tolerance achieved by an autonomous, fully embedded evolvable hardware system, which uses a combination of partial dynamic reconfiguration and an evolutionary algorithm (EA). It demonstrates that the system may self-recover from both transient and cumulative permanent faults. This self-adaptive system, based on a 2D array of 16 (4×4) Processing Elements (PEs), is tested with an image filtering application. Results show that it may properly recover from faults in up to 3 PEs, that is, more than 18% cumulative permanent faults. Two fault models are used for testing purposes, at PE and CLB levels. Two self-healing strategies are also introduced, depending on whether fault diagnosis is available or not. They are based on scrubbing, fitness evaluation, dynamic partial reconfiguration and in-system evolutionary adaptation. Since most of these adaptability features are already available on the system for its normal operation, resource cost for self-healing is very low (only some code additions in the internal microprocessor core)

Automatic Generation of Identical Routing Pairs for FPGA Implemented DPL Logic

Side Channel Attacks (SCAs) typically gather unintentional (side channel) physical leakages from running crypto-devices to reveal confidential data. Dual-rail Precharge Logic (DPL) is one of the most efficient countermeasures against power or EM side channel threats. This logic relies on the implementation of complementary rails to counterbalance the data-dependent variations of the leakage from dynamic behavior of the original circuit. However, the lack of flexibility of commercial FPGA design tools makes it quite difficult to obtain completely balanced routings between complementary networks. In this paper, a controllable repair mechanism to guarantee identical net pairs from two lines is presented: i. repairs the identical yet conflict nets after the duplication (copy & paste) from original rail to complementary rail, and ii. repairs the non-identical nets in off-the-stock DPL circuits; These rerouting steps are carried out starting from a placed and routed netlist using Xilinx Description Language (XDL). Low level XDL modifications have been completely automated using a set of APIs named RapidSmith. Experimental EM attacks show that the resistance level of an AES core after the automatic routing repair is increased in a factor of at least 3.5. Timing analyses further demonstrate that net delay differences between complementary networks are minimized significantly.