Algorithm-oriented design of efficient many-core architectures applied to dense matrix multiplication

Recent integrated circuit technologies have opened the possibility to design parallel architectures with hundreds of cores on a single chip. The design space of these parallel architectures is huge with many architectural options. Exploring the design space gets even more difficult if, beyond performance and area, we also consider extra metrics like performance and area efficiency, where the designer tries to design the architecture with the best performance per chip area and the best sustainable performance. In this paper we present an algorithm-oriented approach to design a many-core architecture. Instead of doing the design space exploration of the many core architecture based on the experimental execution results of a particular benchmark of algorithms, our approach is to make a formal analysis of the algorithms considering the main architectural aspects and to determine how each particular architectural aspect is related to the performance of the architecture when running an algorithm or set of algorithms. The architectural aspects considered include the number of cores, the local memory available in each core, the communication bandwidth between the many-core architecture and the external memory and the memory hierarchy. To exemplify the approach we did a theoretical analysis of a dense matrix multiplication algorithm and determined an equation that relates the number of execution cycles with the architectural parameters. Based on this equation a many-core architecture has been designed. The results obtained indicate that a 100 mm(2) integrated circuit design of the proposed architecture, using a 65 nm technology, is able to achieve 464 GFLOPs (double precision floating-point) for a memory bandwidth of 16 GB/s. This corresponds to a performance efficiency of 71 %. Considering a 45 nm technology, a 100 mm(2) chip attains 833 GFLOPs which corresponds to 84 % of peak performance These figures are better than those obtained by previous many-core architectures, except for the area efficiency which is limited by the lower memory bandwidth considered. The results achieved are also better than those of previous state-of-the-art many-cores architectures designed specifically to achieve high performance for matrix multiplication.

Seasonal patterns of polycyclic aromatic hydrocarbons in digestive gland and arm of octopus (Octopus vulgaris) from the Northwest Atlantic

Among organic pollutants existing in coastal areas, polycyclic aromatic hydrocarbons (PAHs) are of great concern due to their ubiquity and carcinogenic potential. The aim of this study was to evaluate the seasonal patterns of PAHs in the digestive gland and arm of the common octopus (Octopus vulgaris) from the Northwest Atlantic Portuguese coast. In the different seasons, 18 PAHs were determined and the detoxification capacity of the species was evaluated. Ethoxyresorufin O-deethylase (EROD) and ethoxycoumarin O-deethylase (ECOD) activities were measured to assess phase I biotransformation capacity. Individual PAH ratios were used for major source (pyrolytic/petrogenic) analysis. Risks for human consumption were determined by the total toxicity equivalence approach. Generally, low levels of PAHs were detected in the digestive gland and in the arm of octopus, with a predominance of low molecular over high molecular weight compounds. PAHs exhibited seasonality in the concentrations detected and in their main emission sources. In the digestive gland, the highest total PAH levels were observed in autumn possibly related to fat availability in the ecosystem and food intake. The lack of PAH elimination observed in the digestive gland after captivity could be possibly associated to a low biotransformation capacity, consistent with the negligible/undetected levels of EROD and ECOD activity in the different seasons. The emission sources of PAHs found in the digestive gland varied from a petrogenic profile observed in winter to a pyrolytic pattern in spring. In the arm, the highest PAH contents were observed in June; nevertheless, levels were always below the regulatory limits established for food consumption. The carcinogenic potential calculated for all the sampling periods in the arm were markedly lower than the ones found in various aquatic species from different marine environments. The results presented in this study give relevant baseline data for environmental monitoring of organic pollution in coastal areas.

A many-core co-processor for embedded parallel computing on FPGA

Single processor architectures are unable to provide the required performance of high performance embedded systems. Parallel processing based on general-purpose processors can achieve these performances with a considerable increase of required resources. However, in many cases, simplified optimized parallel cores can be used instead of general-purpose processors achieving better performance at lower resource utilization. In this paper, we propose a configurable many-core architecture to serve as a co-processor for high-performance embedded computing on Field-Programmable Gate Arrays. The architecture consists of an array of configurable simple cores with support for floating-point operations interconnected with a configurable interconnection network. For each core it is possible to configure the size of the internal memory, the supported operations and number of interfacing ports. The architecture was tested in a ZYNQ-7020 FPGA in the execution of several parallel algorithms. The results show that the proposed many-core architecture achieves better performance than that achieved with a parallel generalpurpose processor and that up to 32 floating-point cores can be implemented in a ZYNQ-7020 SoC FPGA.

Sparse matrix multiplication on a reconfigurable many-core architecture

Sparse matrix-vector multiplication (SMVM) is a fundamental operation in many scientific and engineering applications. In many cases sparse matrices have thousands of rows and columns where most of the entries are zero, while non-zero data is spread over the matrix. This sparsity of data locality reduces the effectiveness of data cache in general-purpose processors quite reducing their performance efficiency when compared to what is achieved with dense matrix multiplication. In this paper, we propose a parallel processing solution for SMVM in a many-core architecture. The architecture is tested with known benchmarks using a ZYNQ-7020 FPGA. The architecture is scalable in the number of core elements and limited only by the available memory bandwidth. It achieves performance efficiencies up to almost 70% and better performances than previous FPGA designs.

A framework for memory contention analysis in multi-core platforms

The last decade has witnessed a major shift towards the deployment of embedded applications on multi-core platforms. However, real-time applications have not been able to fully benefit from this transition, as the computational gains offered by multi-cores are often offset by performance degradation due to shared resources, such as main memory. To efficiently use multi-core platforms for real-time systems, it is hence essential to tightly bound the interference when accessing shared resources. Although there has been much recent work in this area, a remaining key problem is to address the diversity of memory arbiters in the analysis to make it applicable to a wide range of systems. This work handles diverse arbiters by proposing a general framework to compute the maximum interference caused by the shared memory bus and its impact on the execution time of the tasks running on the cores, considering different bus arbiters. Our novel approach clearly demarcates the arbiter-dependent and independent stages in the analysis of these upper bounds. The arbiter-dependent phase takes the arbiter and the task memory-traffic pattern as inputs and produces a model of the availability of the bus to a given task. Then, based on the availability of the bus, the arbiter-independent phase determines the worst-case request-release scenario that maximizes the interference experienced by the tasks due to the contention for the bus. We show that the framework addresses the diversity problem by applying it to a memory bus shared by a fixed-priority arbiter, a time-division multiplexing (TDM) arbiter, and an unspecified work-conserving arbiter using applications from the MediaBench test suite. We also experimentally evaluate the quality of the analysis by comparison with a state-of-the-art TDM analysis approach and consistently showing a considerable reduction in maximum interference.

Hard real-time multiprocessor scheduling resilient to core failures

Presented at 21st IEEE International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA 2015). 19 to 21, Aug, 2015, pp 122-131. Hong Kong, China.

Polycyclic aromatic hydrocarbons: levels and phase distributions in preschool microenvironment

This work aims to characterize levels and phase distribution of polycyclic aromatic hydrocarbons (PAHs) in indoor air of preschool environment and to assess the impact of outdoor PAH emissions to indoor environment. Gaseous and particulate (PM1 and PM2.5) PAHs (16 USEPA priority pollutants, plus dibenzo[a,l]pyrene, and benzo[j]fluoranthene) were concurrently sampled indoors and outdoors in one urban preschool located in north of Portugal for 35 days. The total concentration of 18 PAHs (ΣPAHs) in indoor air ranged from 19.5 to 82.0 ng/m3; gaseous compounds (range of 14.1–66.1 ng/m3) accounted for 85% ΣPAHs. Particulate PAHs (range 0.7–15.9 ng/m3) were predominantly associated with PM1 (76% particulate ΣPAHs) with 5-ring PAHs being the most abundant. Mean indoor/outdoor ratios (I/O) of individual PAHs indicated that outdoor emissions significantly contributed to PAH indoors; emissions from motor vehicles and fuel burning were the major sources.

Parallel Software framework for time-critical many-core systems

The recent technological advancements and market trends are causing an interesting phenomenon towards the convergence of High-Performance Computing (HPC) and Embedded Computing (EC) domains. On one side, new kinds of HPC applications are being required by markets needing huge amounts of information to be processed within a bounded amount of time. On the other side, EC systems are increasingly concerned with providing higher performance in real-time, challenging the performance capabilities of current architectures. The advent of next-generation many-core embedded platforms has the chance of intercepting this converging need for predictable high-performance, allowing HPC and EC applications to be executed on efficient and powerful heterogeneous architectures integrating general-purpose processors with many-core computing fabrics. To this end, it is of paramount importance to develop new techniques for exploiting the massively parallel computation capabilities of such platforms in a predictable way. P-SOCRATES will tackle this important challenge by merging leading research groups from the HPC and EC communities. The time-criticality and parallelisation challenges common to both areas will be addressed by proposing an integrated framework for executing workload-intensive applications with real-time requirements on top of next-generation commercial-off-the-shelf (COTS) platforms based on many-core accelerated architectures. The project will investigate new HPC techniques that fulfil real-time requirements. The main sources of indeterminism will be identified, proposing efficient mapping and scheduling algorithms, along with the associated timing and schedulability analysis, to guarantee the real-time and performance requirements of the applications.

Influência de um programa de treino da musculatura do core em jogadoras ado-lescentes de voleibol

Introdução: Apesar de na reabilitação e nas atividades da vida diária já se perceber que o treino do core tem bastantes benefícios, no desporto ainda não há provas conclusivas deste mesmo benefício e de como o treino do core deve ser elaborado. Objectivo(s): Avaliar a eficácia de um programa de intervenção sobre a musculatura do core em jogadoras de voleibol Métodos: O presente estudo quasi experimental longitudinal teve uma amostra final de 56 indivíduos – grupo controlo (GC) (n= 27), e grupo experimental (GE) (n=29). Avaliou-se os testes de endurance, e respectivos rácios, descritos por McGill. Ambos os grupos foram avaliados antes (Momento 0) e após (Momento 1) a aplicação de um programa de intervenção. Resultados: Na análise entre os 2 grupos, tanto no Momento 0 como no Momento 1, não se registou diferenças estatisticamente significativas, nem nas médias de tempo dos testes de McGill nem nos rácios. Já na análise intragrupo, o GC não apresentou resultados estatisti-camente significativos, nem nas médias de tempo dos testes do McGill nem nos rácios, enquanto o GE apenas apresentou valores estatisticamente significativos nas médias de tempo da prancha lateral direita (p = 0,04) e prancha lateral esquerda (p = 0,03). Conclusão: Atletas jovens e do sexo feminino, apresentaram um fraco de-sempenho da musculatura extensora do tronco que as torna mais passíveis de contrair lesões lumbopélvicas. Será necessário implementar um programa de intervenção, diferente do aplicado neste estudo, e com maior ênfase na musculatura extensora para equilibrar os rácios propostos por McGill, e assim, diminuir o risco de lesão.

Dominant and Recessive RYR1 Mutations in Adults with Core Lesions and Mild Muscle Symptoms

INTRODUCTION: Ryanodine receptor gene (RYR1) mutations have been associated with central core disease (CCD), multiminicore/minicore/multicore disease (MmD), and susceptibility to malignant hyperthermia (MH). METHODS: Patients with muscle symptoms in adulthood, who had features compatible with CCD/MmD, underwent clinical, histological, and genetic (RYR1 and SEPN1 genes) evaluations. Published cases of CCD and MmD with adult onset were also reviewed. RESULTS: Eight patients fulfilled the criteria for further analysis. Five RYR1 mutations, 4 of them unreported, were detected in 3 patients. Compound heterozygosity was proven in 1 case. CONCLUSIONS: To our knowledge, this is the only report of adult onset associated with recessive RYR1 mutations and central core/multiminicores on muscle biopsy. Although adult patients with CCD, MmD, and minimally symptomatic MH with abnormal muscle biopsy findings usually have a mild clinical course, differential diagnosis and carrier screening is crucial for prevention of potentially life-threatening reactions to general anesthesia.

Investing beyond the core: The case of Soares da Costa

This study aims to analyze how Grupo Soares da Costa, a diversified group centered on construction, behaved in terms of strategy to the current crisis. More specifically, it purposes to understand why Soares da Costa was forced to abandon its strategic plan “Ambições Renovadas”, which was about diversification and internationalization, to decide to focus on it core business. This study uses a SWOT analysis, the examination of the strategic plans and annual reports and the conclusions of two interviews that were carried out. Being the construction sector such a traditional and significant sector to the Portuguese economy, it is important to understand what a company can do to overcome such circumstances. To deal with all the negative circumstances, Soares da Costa should give priority to projects that require low levels of initial capital and diversify geographically to markets with similar characteristics of Angola and Mozambique, where Soares da Costa already excels.

Implementation of Multi-Core Processor Based on PLASMA (Most MIPS I) IP Core

Multi-core processors is a design philosophy that has become mainstream in scientific and engineering applications. Increasing performance and gate capacity of recent FPGA devices has permitted complex logic systems to be implemented on a single programmable device. By using VHDL here we present an implementation of one multi-core processor by using the PLASMA IP core based on the (most) MIPS I ISA and give an overview of the processor architecture and share theexecution results.

Generation of polycyclic aromatic hydrocarbons (PAHs) during woodworking operations.

Occupational exposures to wood dust have been associated with an elevated risk of sinonasal cancer (SNC). Wood dust is recognized as a human carcinogen but the specific cancer causative agent remains unknown. One possible explanation is a co-exposure to; wood dust and polycyclic aromatic hydrocarbons (PAHs). PAHs could be generated during incomplete combustion of wood due to heat created by use of power tools. To determine if PAHs are generated from wood during common wood working operations, PAH concentrations in wood dust samples collected in an experimental chamber operated under controlled conditions were analyzed. In addition, personal air samples from workers exposed to wood dust (n = 30) were collected. Wood dust was generated using three different power tools: vibrating sander, belt sander, and saw; and six wood materials: fir, Medium Density Fiberboard (MDF), beech, mahogany, oak and wood melamine. Monitoring of wood workers was carried out by means of personal sampler device during wood working operations. We measured 21 PAH concentrations in wood dust samples by capillary gas chromatography-ion trap mass spectrometry (GC-MS). Total PAH concentrations in wood dust varied greatly (0.24-7.95 ppm) with the lowest being in MDF dust and the highest in wood melamine dust. Personal PAH exposures were between 37.5-119.8 ng m(-3) during wood working operations. Our results suggest that PAH exposures are present during woodworking operations and hence could play a role in the mechanism of cancer induction related to wood dust exposure.

Occupational exposure of truck drivers to dust and polynuclear aromatic hydrocarbons: a pilot study in Geneva, Switzerland.

The exposure to dust and polynuclear aromatic hydrocarbons (PAH) of 15 truck drivers from Geneva, Switzerland, was measured. The drivers were divided between "long-distance" drivers and "local" drivers and between smokers and nonsmokers and were compared with a control group of 6 office workers who were also divided into smokers and nonsmokers. Dust was measured on 1 workday both by a direct-reading instrument and by sampling. The local drivers showed higher exposure to dust (0.3 mg/m3) and PAH than the long-distance drivers (0.1 mg/m3), who showed no difference with the control group. This observation may be due to the fact that the local drivers spend more time in more polluted areas, such as streets with heavy traffic and construction sites, than do the long-distance drivers. Smoking does not influence exposure to dust and PAH of professional truck drivers, as measured in this study, probably because the ventilation rate of the truck cabins is relatively high even during cold days (11-15 r/h). The distribution of dust concentrations was shown in some cases to be quite different from the expected log-normal distribution. The contribution of diesel exhaust to these exposures could not be estimated since no specific tracer was used. However, the relatively low level of dust exposure dose not support the hypothesis that present day levels of diesel exhaust particulates play a significant role in the excess occurrence of lung cancer observed in professional truck drivers.