The Paralax Infrastructure: Automatic Parallelization With a Helping Hand

Speeding up sequential programs on multicores is a challenging problem that is in urgent need of a solution. Automatic parallelization of irregular pointer-intensive codes, exempli?ed by the SPECint codes, is a very hard problem. This paper shows that, with a helping hand, such auto-parallelization is possible and fruitful. This paper makes the following contributions: (i) A compiler framework for extracting pipeline-like parallelism from outer program loops is presented. (ii) Using a light-weight programming model based on annotations, the programmer helps the compiler to ?nd thread-level parallelism. Each of the annotations speci?es only a small piece of semantic information that compiler analysis misses, e.g. stating that a variable is dead at a certain program point. The annotations are designed such that correctness is easily veri?ed. Furthermore, we present a tool for suggesting annotations to the programmer. (iii) The methodology is applied to autoparallelize several SPECint benchmarks. For the benchmark with most parallelism (hmmer), we obtain a scalable 7-fold speedup on an AMD quad-core dual processor. The annotations constitute a parallel programming model that relies extensively on a sequential program representation. Hereby, the complexity of debugging is not increased and it does not obscure the source code. These properties could prove valuable to increase the ef?ciency of parallel programming.

Towards a better understanding of the nature and contribution of the interflow to catchment dynamics.

A detailed understanding of flow and contaminant transfer along each of the key hydrological pathways within a catchment is critical for designing and implementing cost effective Programmes of Measures under the Water
Framework Directive.
The Contaminant Movement along Pathways Project (’The Pathways Project’) is an Irish, EPA STRIVE funded, large multi-disciplinary project which is focussed on understanding and modelling flow and attenuation along each of these pathways for the purposes of developing a catchment management tool. The tool will be used by EPA and RBD catchment managers to assess and manage the impacts of diffuse contamination on stream aquatic ecology. Four main contaminants of interest — nitrogen, phosphorus, sediment and pathogens — are being
investigated in four instrumented test catchments. In addition to the usual hydrological and water chemistry/quality parameters typically captured in catchment studies, field measurements at the test catchments include ecological
sampling, sediment dynamics, soil moisture dynamics, and groundwater levels and chemistry/quality, both during and between significant rainfall events. Spatial and temporal sampling of waters directly from the pathways of
interest is also being carried out.
Sixty-five percent of Ireland is underlain by poorly productive aquifers. In these hydrogeological settings, the main pathways delivering flow to streams are overland flow, interflow and shallow bedrock flow. Little is
known about the interflow pathway and its relative importance in delivery of flow and contaminants to the streams. Interflow can occur in both the topsoil and subsoil, and may include unsaturated matrix flow, bypass or macropore
flow, saturated flow in locally perched water tables and artificial field drainage.
Results to date from the test catchment experiments show that artificial field drains play an important role in the delivery of interflow to these streams, during and between rainfall events when antecedent conditions are
favourable. Hydrochemical mixing models, using silica and SAC254 (the absorbance of UV light at a wavelength of 254 nm which is a proxy for dissolved organic matter) as tracers, show that drain flow is an important end
member contributing to the stream and that proportionally, its contribution is relatively high.
Results from the study also demonstrate that waters originating from one pathway often mix with the waters from another, and are subsequently delivered to the stream at rates, and with chemical/quality characteristics,
that are not typical of either pathway. For example, pre-event shallow groundwater not far from the catchment divide comes up to the surface as rejected recharge during rainfall events and is rapidly delivered to the stream
via overland flow and/or artificial land drainage, bringing with it higher nitrate than would often be expected from a quickflow pathway contribution. This is contrary to the assumption often made in catchment studies that the
deeper hydrological pathways have slower response times in stream hydrographs during a rainfall event, and it emphasizes that it is critical to have a strong three-dimensional conceptual model as the basis for the interpretation
of catchment data.

Integrating geophysical data and upscaling techniques to support regional groundwater flow modelling: A practical example of the Lagan Valley, Northern Ireland

When studying heterogeneous aquifer systems, especially at regional scale, a degree of generalization is anticipated. This can be due to sparse sampling regimes, complex depositional environments or lack of accessibility to measure the subsurface. This can lead to an inaccurate conceptualization which can be detrimental when applied to groundwater flow models. It is important that numerical models are based on observed and accurate geological information and do not rely on the distribution of artificial aquifer properties. This can still be problematic as data will be modelled at a different scale to which it was collected. It is proposed here that integrating geophysics and upscaling techniques can assist in a more realistic and deterministic groundwater flow model. In this study, the sedimentary aquifer of the Lagan Valley in Northern Ireland is chosen due to intruding sub-vertical dolerite dykes. These dykes are of a lower permeability than the sandstone aquifer. The use of airborne magnetics allows the delineation of heterogeneities, confirmed by field analysis. Permeability measured at the field scale is then upscaled to different levels using a correlation with the geophysical data, creating equivalent parameters that can be directly imported into numerical groundwater flow models. These parameters include directional equivalent permeabilities and anisotropy. Several stages of upscaling are modelled in finite element. Initial modelling is providing promising results, especially at the intermediate scale, suggesting an accurate distribution of aquifer properties. This deterministic based methodology is being expanded to include stochastic methods of obtaining heterogeneity location based on airborne geophysical data. This is through the Direct Sample method of Multiple-Point Statistics (MPS). This method uses the magnetics as a training image to computationally determine a probabilistic occurrence of heterogeneity. There is also a need to apply the method to alternate geological contexts where the heterogeneity is of a higher permeability than the host rock.

Assessing the environmental fate of selected polybrominated diphenyl ethers in the region surrounding the Zhuoshui River of Taiwan based on an Equilibrium Constant fugacity model

Polybrominated diphenyl ethers (PBDEs) are a group of flame retardants that have been in use since the 1970s. They are included in the list of hazardous substances known as persistent organic pollutants (POPs) because they are extremely hazardous to the environment and human health. PBDEs have been extensively used in industry and manufacturing in Taiwan, thus its citizens are at high risk of exposure to these chemicals.
An assessment of the environmental fate of these compounds in the Zhuoshui river and Changhua County regions of western Taiwan, and also including the adjacent area of the Taiwan Strait, was conducted for three high risk congeners, BDE-47, -99 and -209, to obtain information regarding the partitioning, advection, transfer and long range transport potential of the PBDEs in order to identify the level of risk posed by the pollutants in this region.
The results indicate that large amounts of PBDEs presently reside in all model compartments – air, soil, water, and sediment – with particularly high levels found in air and especially in sediment. The high levels found in sediment, particularly for BDE-209, are significant, since there is the threat of these pollutants entering the food chain, either directly through benthic feeding, or through resuspension and subsequent feeding in the pelagic region of the water column which is a distinct possibility in the strong currents found within the Taiwan Strait. Another important result is that a substantial portion of emissions leave the model domain directly through advection, particularly for BDE-47 (58%) and BDE-209 (75%), thus posing a risk to adjacent communities.
Model results were generally in reasonable agreement with available measured concentrations. In air, model concentrations are in reasonably good agreement with available measured values. For both BDE-47 and -99, model concentrations are a factor of 2-3 higher and BDE-209 within the range of measured values. In soil, model results are somewhat less than measured values. In sediment, model results are at the high end of measured values.