A cluster-based implementation of a fault-tolerant parallel reduction algorithm using swarm-array computing

Recent research in multi-agent systems incorporate fault tolerance concepts. However, the research does not explore the extension and implementation of such ideas for large scale parallel computing systems. The work reported in this paper investigates a swarm array computing approach, namely ‘Intelligent Agents’. In the approach considered a task to be executed on a parallel computing system is decomposed to sub-tasks and mapped onto agents that traverse an abstracted hardware layer. The agents intercommunicate across processors to share information during the event of a predicted core/processor failure and for successfully completing the task. The agents hence contribute towards fault tolerance and towards building reliable systems. The feasibility of the approach is validated by simulations on an FPGA using a multi-agent simulator and implementation of a parallel reduction algorithm on a computer cluster using the Message Passing Interface.

Ventilation for air quality and energy efficiency

This article addresses the need for providing good standards of indoor air quality (IAQ) in buildings from the view point of health, well-being and productivity of building occupants. It briefly outlines the role of ventilation in achieving the required IAQ targets and discusses the performance of different types of ventilation systems in use. As a result of new energy efficiency directives and legislations in Europe and elsewhere, the ventilation energy component of HVAC systems has increased in relative terms and this article introduces a method for evaluating the performance air distribution systems that is based on ventilation and energy effectiveness. A range of ventilation systems are discussed, including mechanical and natural ventilation, and results for more recently developed mechanical air distribution systems are compared with conventional systems. The article provides an assessment and comparison of some of these systems with reference to ventilation performance and energy efficiency

Biomass partitioning and growth efficiency in four naturally regenerated forest tree species

Current forest growth models and yield tables are almost exclusively based on data from mature trees, reducing their applicability to young and developing stands. To address this gap, young European beech, sessile oak, Scots pine and Norway spruce trees approximately 0 to 10 years old were destructively sampled in a range of naturally regenerated forest stands in Central Europe. Diameter at base and height were first measured in situ for up to 175 individuals per species. Subsequently, the trees were excavated and dry biomass of foliage, branches, stems and roots was measured. Allometric relations were then used to calculate biomass allocation coefficients (BAC) and growth efficiency (GE) patterns in young trees. We found large differences in BAC and GE between broadleaves and conifers, but also between species within these categories. Both BAC and GE are strongly age-specific in young trees, their rapidly changing values reflecting different growth strategies in the earliest stages of growth. We show that linear relationships describing biomass allocation in older trees are not applicable in young trees. To accurately predict forest biomass and carbon stocks, forest growth models need to include species and age specific parameters of biomass allocation patterns.

Neurofuzzy mixture of experts network parallel learning and model construction algorithms

A connection between a fuzzy neural network model with the mixture of experts network (MEN) modelling approach is established. Based on this linkage, two new neuro-fuzzy MEN construction algorithms are proposed to overcome the curse of dimensionality that is inherent in the majority of associative memory networks and/or other rule based systems. The first construction algorithm employs a function selection manager module in an MEN system. The second construction algorithm is based on a new parallel learning algorithm in which each model rule is trained independently, for which the parameter convergence property of the new learning method is established. As with the first approach, an expert selection criterion is utilised in this algorithm. These two construction methods are equivalent in their effectiveness in overcoming the curse of dimensionality by reducing the dimensionality of the regression vector, but the latter has the additional computational advantage of parallel processing. The proposed algorithms are analysed for effectiveness followed by numerical examples to illustrate their efficacy for some difficult data based modelling problems.

Efficiency and purity control in the preparation of pure and/or aluminum-doped barium ferrites by hydrothermal methods using ferrous ions as reactants

The synthesis of hexagonal barium ferrite (BaFe12O19) was studied under hydrothermal conditions by a method in which a significant amount of ferrous chloride was introduced along side ferric chloride among the starting materials. Though all of the Fe2+ ions in the starting material were converted to Fe3+ ions in the final product, Fe2+ was confirmed to participate differently from the Fe3+ used in the conventional method in the mechanism of forming barium ferrite. Indeed the efficiency of the synthesis and the quality of the product and the lack of impurities such as Fe2O3 and BaFe2O4 were improved when Fe2+ was included. However, the amount of ferrous ions that could be included to obtain the desired product was limited with an optimum ratio of 2:8 for FeCl2/FeCl3 when only 2h of reaction time were needed. It was also found that the role of trivalent Fe3+ could be successfully replaced by Al3+. Up to 50% of their on could be replaced by Al3+ in the reactants to produce Al- doped products. It was also found that the ratio of Fe2+/M3+ could be increased in the presence of Al3+ to produce high quality barium ferrite.

Farm productivity and efficiency in rural Bangladesh: the role of education revisited

This article reassesses the debate over the role of education in farm production in Bangladesh using a large dataset on rice producing households from 141 villages. Average and stochastic production frontier functions are estimated to ascertain the effect of education on productivity and efficiency. A full set of proxies for farm education stock variables are incorporated to investigate the ‘internal’ as well as ‘external’ returns to education. The external effect is investigated in the context of rural neighbourhoods. Our analysis reveals that in addition to raising rice productivity and boosting potential output, household education significantly reduces production inefficiencies. However, we are unable to find any evidence of the externality benefit of schooling – neighbour's education does not matter in farm production. We discuss the implication of these findings for rural education programmes in Bangladesh.

Effect of wheat dwarfing genes on nitrogen-use efficiency

Near isogenic lines (NILs) varying for alleles for reduced height (Rht) and photoperiod insensitivity (Ppd-D1a) in a cvar Mercia background (rht (tall), Rht-B1b, Rht-D1b, Rht-B1c, Rht8c+Ppd-D1a, Rht-D1c, Rht12) were compared at a field site in Berkshire, UK, but within different systems (‘organic’, O, in 2005/06, 2006/07 and 2007/08 growing seasons v. ‘conventional’, C, in 2005/06, 2006/07, 2007/08 and 2008/09). In 2007 and 2008, further NILs (rht (tall), Rht-B1b, Rht-D1b, Rht-B1c, Rht-B1b+Rht-D1b, Rht-D1b+Rht-B1c) in both Maris Huntsman and Maris Widgeon backgrounds were added. The contrasting systems allowed NILs to be tested in diverse rotational and agronomic, but commercially relevant, contexts, particularly with regard to the assumed temporal distribution of nitrogen availability, and competition from weeds. For grain, nitrogen-use efficiency (NUE; grain dry matter (DM) yield/available N; where available N=fertilizer N+soil mineral N), recovery of N in the grain (grain N yield/available N), N utilization efficiency to produce grain (NUtEg; grain DM yield/above-ground crop N yield), N harvest index (grain N yield/above-ground crop N yield) and dry matter harvest index (DMHI; grain DM yield/above-ground crop DM yield) all peaked at final crop heights of 800–950 mm. Maximum NUE occurred at greater crop heights in the organic system than in the conventional system, such that even adding just a semi-dwarfing allele (Rht-D1b) to the shortest background, Mercia, reduced NUE in the organic system. The mechanism of dwarfing (gibberellin sensitive or insensitive) made little difference to the relationship between NUE and its components with crop height. For above-ground biomass: dwarfing alleles had a greater effect on DM accumulation compared with N accumulation such that all dwarfing alleles could reduce nitrogen utilization efficiency (NUtE; crop DM yield/crop N yield). This was particularly evident at anthesis in the conventional system when there was no significant penalty for severe dwarfism for N accumulation, despite a 3-tonne (t)/ha reduction in biomass compared to the tallest lines. Differences between genotypes for recovery of N in the grain were thus mostly a function of net N uptake after anthesis rather than of remobilized N. This effect was compounded as dwarfing, except when coupled with Ppd-D1a, was associated with delayed anthesis. In the organic experiments there was greater reliance on N accumulated before anthesis, and genotype effects on NUE were confounded with effects on N accumulated by weeds, which was negatively associated with crop height. Optimum height for maximizing wheat NUE and its components, as manipulated by Rht alleles, thus depend on growing system, and crop utilization (i.e. biomass or grain production).