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.

High resolution stratigraphy and miocene facies correlation in Lisbon and Setúbal Peninsula (Lower Tagus basin, Portugal)

Correlation between facies associations (marine, estuarine and distal fluviatile environments) and disconformities, observed between Foz da Fonte (SW of Setúbal Peninsula) and Santa Iria da Azóia (NE of Lisbon) are presented. The precise definition of the marine-continental facies relationships improved very much the chronology of the depositional sequence boundaries. Tectonic and eustatic controls are discussed on the basis of subsidence rates variation.

High-resolution correlation of coastal cliff sections in the Lagos- Portimao Formation (Lower - Middle Miocene, central Algarve, Portugal)

This paper describes a high-resolution stratigraphic correlation scheme for the early to middle Miocene Lagos-Portimão Formation of central Algarve, southern Portugal. The Lagos Portimão-Formation of central Algarve is a 60 m thick package of horizontally bedded siliciclastics and carbonates. The bryozoan and mollusc dominated biofacies is typical of a shallow marine, warm-temperate climatic environment. We define four stratigraphic marker beds based on biofacies, lithology, and gamma-ray signatures. Marker bed 1 is a reddish shell bed composed predominantly of bivalve shells in various stages of fragmentation. Marker bed 2 is a fossiliferous sandstone / sandy rudstone characterized by bryozoan masses. Marker bed 3 is also a fossiliferous sandstone with abundant larger foraminifers and foliate bryozoans. Marker bed 4 is composed of three distinct layers; two fossiliferous sandstones with an intercalated shell bed. The upper sandstone unit displays thickets of the bryozoan Celleporaria palmate associated with the coral Culizia parasitica. This stratigraphic framework allows to correlate isolated outcrops within the stratigraphic context of the Lagos-Portimão Formation and to establish high resolution chronostratigraphic Sr-isotopic dating.

Matrix converter as unified power flow controller: design and implementation of decoupled direct power controllers

This paper presents the design and implementation of direct power controllers for three-phase matrix converters (MC) operating as Unified Power Flow Controllers (UPFC). Theoretical principles of the decoupled linear power controllers of the MC-UPFC to minimize the cross-coupling between active and reactive power control are established. From the matrix converter based UPFC model with a modified Venturini high frequency PWM modulator, decoupled controllers for the transmission line active (P) and reactive (Q) power direct control are synthesized. Simulation results, obtained from Matlab/Simulink, are presented in order to confirm the proposed approach. Results obtained show decoupled power control, zero error tracking, and fast responses with no overshoot and no steady-state error.

On derivatives and norms of generalized matrix functions and respective symmetric powers

In recent papers, the authors obtained formulas for directional derivatives of all orders, of the immanant and of the m-th xi-symmetric tensor power of an operator and a matrix, when xi is a character of the full symmetric group. The operator norm of these derivatives was also calculated. In this paper, similar results are established for generalized matrix functions and for every symmetric tensor power.

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.

Estrogen receptor dependent genetic and epigenetic factors of tamoxifen resistance

Resumo: A decisão da terapêutica hormonal no tratamento do cancro da mama baseiase na determinação do receptor de estrogénio alfa por imunohistoquímica (IHC). Contudo, a presença deste receptor não prediz a resposta em todas as situações, em parte devido a limitações do método IHC. Investigámos se a expressão dos genes ESR1 e ESR2, bem como a metilação dos respectivos promotores, pode estar relacionada com a evolução desfavorável de uma proporção de doentes tratados com tamoxifeno assim como com a perda dos receptores de estrogénio alfa (ERα) e beta (ERß). Amostras de 211 doentes com cancro da mama diagnosticado entre 1988 e 2004, fixadas em formalina e preservadas em parafina, foram utilizadas para a determinação por IHC da presença dos receptores ERα e ERß. O mRNA total do gene ESR1 e os níveis específicos do transcrito derivado do promotor C (ESR1_C), bem como dos transcritos ESR2_ß1, ESR2_ß2/cx, and ESR2_ß5 foram avaliados por Real-time PCR. Os promotores A e C do gene ESR1 e os promotores 0K e 0N do gene ESR2 foram investigados por análise de metilação dos dinucleotidos CpG usando bisulfite-PCR para análise com enzimas de restrição, ou para methylation specific PCR. Atendendo aos resultados promissores relacionados com a metilação do promotor do gene ESR1, complementamos o estudo com um método quantitativo por matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) suportado pelo software Epityper para a medição da metilação nos promotores A e C. Fez-se a avaliação da estabilidade do mRNA nas linhas celulares de cancro da mama MCF-7 e MDA-MB-231 tratadas com actinomicina D. Baixos níveis do transcrito ESR1_C associaram-se a uma melhor sobrevivência global (p = 0.017). Níveis elevados do transcrito ESR1_C associaram-se a uma resposta inferior ao tamoxifeno (HR = 2.48; CI 95% 1.24-4.99), um efeito mais pronunciado em doentes com tumores de fenótipo ERα/PgR duplamente positivo (HR = 3.41; CI 95% 1.45-8.04). A isoforma ESR1_C mostrou ter uma semi-vida prolongada, bem como uma estrutura secundária da região 5’UTR muito mais relaxada em comparação com a isoforma ESR1_A. A análise por Western-blot mostrou que ao nível da 21 proteína, a selectividade de promotores é indistinguivel. Não se detectou qualquer correlação entre os níveis das isoformas do gene ESR2 ou entre a metilação dos promotores do gene ESR2, e a detecção da proteína ERß. A metilação do promotor C do gene ESR1, e não do promotor A, foi responsável pela perda do receptor ERα. Estes resultados sugerem que os níveis do transcrito ESR1_C sejam usados como um novo potencial marcador para o prognóstico e predição de resposta ao tratamento com tamoxifeno em doentes com cancro da mama. Abstract: The decision of endocrine breast cancer treatment relies on ERα IHC-based assessment. However, ER positivity does not predict response in all cases in part due to IHC methodological limitations. We investigated whether ESR1 and ESR2 gene expression and respective promoter methylation may be related to non-favorable outcome of a proportion of tamoxifen treated patients as well as to ERα and ERß loss. Formalin-fixed paraffin-embedded breast cancer samples from 211 patients diagnosed between 1988 and 2004 were submitted to IHC-based ERα and ERß protein determination. ESR1 whole mRNA and promoter C specific transcript levels, as well as ESR2_ß1, ESR2_ß2/cx, and ESR2_ß5 transcripts were assessed by real-time PCR. ESR1 promoters A and C, and ESR2 promoters 0N and 0K were investigated by CpG methylation analysis using bisulfite-PCR for restriction analysis, or methylation specific PCR. Due to the promising results related to ESR1 promoter methylation, we have used a quantification method by matrixassisted laser desorption/ionization time-of-flight mass spectrometry (MALDITOF MS) together with Epityper software to measure methylation at promoters A and C. mRNA stability was assessed in actinomycin D treated MCF-7 and MDA-MB-231 cells. ERα protein was quantified using transiently transfected breast cancer cells. Low ESR1_C transcript levels were associated with better overall survival (p = 0.017). High levels of ESR1_C transcript were associated with non-favorable response in tamoxifen treated patients (HR = 2.48; CI 95% 1.24-4.99), an effect that was more pronounced in patients with ERα/PgR double-positive tumors (HR = 3.41; CI 95% 1.45-8.04). The ESR1_C isoform had a prolonged mRNA half-life and a more relaxed 5’UTR structure compared to ESR1_A isoform. Western-blot analysis showed that at protein level, the promoter selectivity is undistinguishable. There was no correlation between levels of ESR2 isoforms or ESR2 promoter methylation and ERß protein staining. ESR1 promoter C CpG methylation and not promoter A was responsible for ERα loss. We propose ESR1_C levels as a putative novel marker for breast cancer prognosis and prediction of tamoxifen response.

Solubilities of C‑Tetraalkylcalix[4]resorcinarenes in SCCO2: experimental measurements, characterization, and correlation

The solubilities of two C-tetraalkylcalix[4]resorcinarenes, namely C-tetramethylcalix[4]resorcinarene and C-tetrapentylcalix[4]resorcinarene, in supercritical carbon dioxide (SCCO2) were measured in a flow-type apparatus at a temperature range from (313.2 to 333.2) K and at pressures from (12.0 to 35.0) MPa. The C-tetraalkylcalix[4]resorcinarenes were synthesized applying our optimized procedure and fully characterized by means of gel permeation chromatography, infrared and nuclear magnetic resonance spectroscopy. The solubilities of the C-tetraalkylcalix[4]resorcinarenes in SCCO2 were determined by analysis of the extracts obtained by HPLC with ultraviolet (UV) detection methodology adapted by our team. Four semiempirical density-based models, and the SoaveRedlichKwong cubic equation of state (SRK CEoS) with classical mixing rules, were applied to correlate the solubility of the calix[4]resorcinarenes in the SC CO2. The physical properties required for the modeling were estimated and reported.

Off-the-person electrocardiography: performance assessment and clinical correlation

Previous work by our group introduced a novel concept and sensor design for “off-the-person” ECG, for which evidence on how it compares against standard clinical-grade equipment has been largely missing. Our objectives with this work are to characterise the off-the-person approach in light of the current ECG systems landscape, and assess how the signals acquired using this simplified setup compare with clinical-grade recordings. Empirical tests have been performed with real-world data collected from a population of 38 control subjects, to analyze the correlation between both approaches. Results show off-the-person data to be correlated with clinical-grade data, demonstrating the viability of this approach to potentially extend preventive medicine practices by enabling the integration of ECG monitoring into multiple dimensions of people’s everyday lives. © 2015, IUPESM and Springer-Verlag Berlin Heidelberg.

Vertex component analysis: a geometric-based approach to unmix hyperspectral data

Hyperspectral remote sensing exploits the electromagnetic scattering patterns of the different materials at specific wavelengths [2, 3]. Hyperspectral sensors have been developed to sample the scattered portion of the electromagnetic spectrum extending from the visible region through the near-infrared and mid-infrared, in hundreds of narrow contiguous bands [4, 5]. The number and variety of potential civilian and military applications of hyperspectral remote sensing is enormous [6, 7]. Very often, the resolution cell corresponding to a single pixel in an image contains several substances (endmembers) [4]. In this situation, the scattered energy is a mixing of the endmember spectra. A challenging task underlying many hyperspectral imagery applications is then decomposing a mixed pixel into a collection of reflectance spectra, called endmember signatures, and the corresponding abundance fractions [8–10]. Depending on the mixing scales at each pixel, the observed mixture is either linear or nonlinear [11, 12]. Linear mixing model holds approximately when the mixing scale is macroscopic [13] and there is negligible interaction among distinct endmembers [3, 14]. If, however, the mixing scale is microscopic (or intimate mixtures) [15, 16] and the incident solar radiation is scattered by the scene through multiple bounces involving several endmembers [17], the linear model is no longer accurate. Linear spectral unmixing has been intensively researched in the last years [9, 10, 12, 18–21]. It considers that a mixed pixel is a linear combination of endmember signatures weighted by the correspondent abundance fractions. Under this model, and assuming that the number of substances and their reflectance spectra are known, hyperspectral unmixing is a linear problem for which many solutions have been proposed (e.g., maximum likelihood estimation [8], spectral signature matching [22], spectral angle mapper [23], subspace projection methods [24,25], and constrained least squares [26]). In most cases, the number of substances and their reflectances are not known and, then, hyperspectral unmixing falls into the class of blind source separation problems [27]. Independent component analysis (ICA) has recently been proposed as a tool to blindly unmix hyperspectral data [28–31]. ICA is based on the assumption of mutually independent sources (abundance fractions), which is not the case of hyperspectral data, since the sum of abundance fractions is constant, implying statistical dependence among them. This dependence compromises ICA applicability to hyperspectral images as shown in Refs. [21, 32]. In fact, ICA finds the endmember signatures by multiplying the spectral vectors with an unmixing matrix, which minimizes the mutual information among sources. If sources are independent, ICA provides the correct unmixing, since the minimum of the mutual information is obtained only when sources are independent. This is no longer true for dependent abundance fractions. Nevertheless, some endmembers may be approximately unmixed. These aspects are addressed in Ref. [33]. Under the linear mixing model, the observations from a scene are in a simplex whose vertices correspond to the endmembers. Several approaches [34–36] have exploited this geometric feature of hyperspectral mixtures [35]. Minimum volume transform (MVT) algorithm [36] determines the simplex of minimum volume containing the data. The method presented in Ref. [37] is also of MVT type but, by introducing the notion of bundles, it takes into account the endmember variability usually present in hyperspectral mixtures. The MVT type approaches are complex from the computational point of view. Usually, these algorithms find in the first place the convex hull defined by the observed data and then fit a minimum volume simplex to it. For example, the gift wrapping algorithm [38] computes the convex hull of n data points in a d-dimensional space with a computational complexity of O(nbd=2cþ1), where bxc is the highest integer lower or equal than x and n is the number of samples. The complexity of the method presented in Ref. [37] is even higher, since the temperature of the simulated annealing algorithm used shall follow a log( ) law [39] to assure convergence (in probability) to the desired solution. Aiming at a lower computational complexity, some algorithms such as the pixel purity index (PPI) [35] and the N-FINDR [40] still find the minimum volume simplex containing the data cloud, but they assume the presence of at least one pure pixel of each endmember in the data. This is a strong requisite that may not hold in some data sets. In any case, these algorithms find the set of most pure pixels in the data. PPI algorithm uses the minimum noise fraction (MNF) [41] as a preprocessing step to reduce dimensionality and to improve the signal-to-noise ratio (SNR). The algorithm then projects every spectral vector onto skewers (large number of random vectors) [35, 42,43]. The points corresponding to extremes, for each skewer direction, are stored. A cumulative account records the number of times each pixel (i.e., a given spectral vector) is found to be an extreme. The pixels with the highest scores are the purest ones. N-FINDR algorithm [40] is based on the fact that in p spectral dimensions, the p-volume defined by a simplex formed by the purest pixels is larger than any other volume defined by any other combination of pixels. This algorithm finds the set of pixels defining the largest volume by inflating a simplex inside the data. ORA SIS [44, 45] is a hyperspectral framework developed by the U.S. Naval Research Laboratory consisting of several algorithms organized in six modules: exemplar selector, adaptative learner, demixer, knowledge base or spectral library, and spatial postrocessor. The first step consists in flat-fielding the spectra. Next, the exemplar selection module is used to select spectral vectors that best represent the smaller convex cone containing the data. The other pixels are rejected when the spectral angle distance (SAD) is less than a given thresh old. The procedure finds the basis for a subspace of a lower dimension using a modified Gram–Schmidt orthogonalizati on. The selected vectors are then projected onto this subspace and a simplex is found by an MV T pro cess. ORA SIS is oriented to real-time target detection from uncrewed air vehicles using hyperspectral data [46]. In this chapter we develop a new algorithm to unmix linear mixtures of endmember spectra. First, the algorithm determines the number of endmembers and the signal subspace using a newly developed concept [47, 48]. Second, the algorithm extracts the most pure pixels present in the data. Unlike other methods, this algorithm is completely automatic and unsupervised. To estimate the number of endmembers and the signal subspace in hyperspectral linear mixtures, the proposed scheme begins by estimating sign al and noise correlation matrices. The latter is based on multiple regression theory. The signal subspace is then identified by selectin g the set of signal eigenvalue s that best represents the data, in the least-square sense [48,49 ], we note, however, that VCA works with projected and with unprojected data. The extraction of the end members exploits two facts: (1) the endmembers are the vertices of a simplex and (2) the affine transformation of a simplex is also a simplex. As PPI and N-FIND R algorithms, VCA also assumes the presence of pure pixels in the data. The algorithm iteratively projects data on to a direction orthogonal to the subspace spanned by the endmembers already determined. The new end member signature corresponds to the extreme of the projection. The algorithm iterates until all end members are exhausted. VCA performs much better than PPI and better than or comparable to N-FI NDR; yet it has a computational complexity between on e and two orders of magnitude lower than N-FINDR. The chapter is structure d as follows. Section 19.2 describes the fundamentals of the proposed method. Section 19.3 and Section 19.4 evaluate the proposed algorithm using simulated and real data, respectively. Section 19.5 presents some concluding remarks.

Viscosity and self-diffusion coefficients of dialkyl adipates: a correlation scheme with predictive capabilities

A correlation and predictive scheme for the viscosity and self-diffusivity of liquid dialkyl adipates is presented. The scheme is based on the kinetic theory for dense hard-sphere fluids, applied to the van der Waals model of a liquid to predict the transport properties. A "universal" curve for a dimensionless viscosity of dialkyl adipates was obtained using recently published experimental viscosity and density data of compressed liquid dimethyl (DMA), dipropyl (DPA), and dibutyl (DBA) adipates. The experimental data are described by the correlation scheme with a root-mean-square deviation of +/- 0.34 %. The parameters describing the temperature dependence of the characteristic volume, V-0, and the roughness parameter, R-eta, for each adipate are well correlated with one single molecular parameter. Recently published experimental self-diffusion coefficients of the same set of liquid dialkyl adipates at atmospheric pressure were correlated using the characteristic volumes obtained from the viscosity data. The roughness factors, R-D, are well correlated with the same single molecular parameter found for viscosity. The root-mean-square deviation of the data from the correlation is less than 1.07 %. Tests are presented in order to assess the capability of the correlation scheme to estimate the viscosity of compressed liquid diethyl adipate (DEA) in a range of temperatures and pressures by comparison with literature data and of its self-diffusivity at atmospheric pressure in a range of temperatures. It is noteworthy that no data for DEA were used to build the correlation scheme. The deviations encountered between predicted and experimental data for the viscosity and self-diffusivity do not exceed 2.0 % and 2.2 %, respectively, which are commensurate with the estimated experimental measurement uncertainty, in both cases.

The effect of water inorganic matrix in ibuprofen adsorption onto activated carbon for water and wastewater treatment

Dissertação para obtenção do Grau de Mestre em Engenharia Química e Bioquímica

Spatial neighborhood matrix computation: Inverse distance weighted versus binary contiguity

Paper presented at Geo-Spatial Crossroad GI_Forum, Salzburg, Austria.