Comparing Processor Allocation Strategies in Multiprogrammed Shared-Memory Multiprocessors

Small-scale shared-memory multiprocessors are commonly used in a workgroup environment where multiple applications, both parallel and sequential, are executed concurrently while sharing the processors and other system resources. To utilize the processors efficiently, an effective allocation strategy is required. In this paper, we use performance data obtained from an SGI multiprocessor to evaluate several processor allocation strategies when running two parallel programs simultaneously. We examine gang scheduling (coscheduling), static space-sharing (space partitioning), and a dynamic allocation scheme called loop-level process control (LLPC) with three different dynamic allocation heuristics. We use regression analysis to quantify the measured data and thereby explore the relationship between the degree of parallelism of the application, specific system parameters (such as the size of the system), the processor allocation strategy, and the resulting performance. This study shows that dynamically partitioning the system using LLPC or similar heuristics provides better performance for applications with a high degree of parallelism than either gang scheduling or static space-sharing.     

Multilevel survival modelling of recurrent urinary tract infections

A multilevel survival frailty model is presented for analyzing clustered and recurrent urinary tract infections among elderly women residing in aged-care institutions. At the subject level, serial dependence is expected between recurrent events recorded on the same individual. At the cluster level, correlations of observations within the same institution are present due to the inherent residential environment and hierarchical setting. Two random components are therefore incorporated explicitly within the survival frailty model to account for the simultaneous heterogeneity and autoregressive structure. A Splus computer program is developed for the estimation of fixed effect and variance component parameters.     

Decolourisation of Orange II by a Wood-Rotting Fungus

The decolourisation of Orange II by a wood-rotting fungus has been studied. It was found that Fungus F29 could effectively decolourise Orange II especially when grown as pelleted mycelia under agitated conditions. Many factors affecting the decolourisation process in nitrogen-limited media (NLM) were studied, including: concentration of glucose, NH₄⁺, Mn(II) and veratryl alcohol; initial pH; amount of mycelium; mycelial age; Orange II concentration; temperature. Results showed that the media containing Orange II at 1000 mg dm⁻³ (or higher) could be decolourised by 98% of the initial colour (A_{480 nm}) in 2 days, in most conditions tested, and that the mycelia could be repeatedly reused. © 1997 SCI.     

Remediation and recovery of uranium from contaminated subsurface environments with electrodes

Previous studies have demonstrated that Geobacter species can effectively remove uranium from contaminated groundwater by reducing soluble U(VI) to the relatively insoluble U(IV) with organic compounds serving as the electron donor. Studies were conducted to determine whether electrodes might serve as an alternative electron donor for U(VI) reduction by a pure culture of Geobacter sulfurreducens and microorganisms in uranium-contaminated sediments. Electrodes poised at -500 mV (vs a Ag/AgCl reference) rapidly removed U(VI) from solution in the absence of cells. However, when the poise at the electrode was removed, all of the U(VI) returned to solution, demonstrating that the electrode did not reduce U(VI). If G. sulfurreducens was present on the electrode, U(VI) did not return to solution until the electrode was exposed to dissolved oxygen. This suggeststhat G. sulfurreducens on the electrode reduced U(VI) to U(IV) which was stably precipitated until reoxidized in the presence of oxygen. When an electrode was placed in uranium-contaminated subsurface sediments, U(VI) was removed and recovered from groundwater using poised electrodes. Electrodes emplaced in flow-through columns of uranium-contaminated sediments readily removed U(VI) from the groundwater, and 87% of the uranium that had been removed was recovered from the electrode surface after the electrode was pulled from the sediments. These results suggest that microorganisms can use electrons derived from electrodes to reduce U(VI) and that it may be possible to remove and recover uranium from contaminated groundwater with poised electrodes.     

Science and Engineering in the Petascale Era

What breakthrough advances will petascale computing bring to various science and engineering fields? Experts in everything from astronomy to seismology envision the opportunities ahead and the impact they'll have on advancing our understanding of the world.     

Fast ray tracing and the potential effects on graphics and gaming courses

The modern graphics processing units (GPUs), found on almost every personal computer, use the z-buffer algorithm to compute visibility. Ray tracing, an alternative to the z-buffer algorithm, delivers higher visual quality than the z-buffer algorithm but has historically been too slow for interactive use. However, ray tracing has benefited from improvements in computer hardware, and many believe it will replace the z-buffer algorithm as the graphics engine on PCs. If this replacement happens, it will imply fundamental changes in both the API to and capabilities of 3D graphics engines. This paper overviews the backgrounds in z-buffer and ray tracing, presents our case that ray tracing will replace z-buffer in the near future, and discusses the implications for graphics oriented classes should this switch to ray tracing occur. Since computer gaming is one of the most important industry driving graphics hardware and the fact that recently there are many computer science courses related to games and games development, we also describe the potential impact on games related classes.     

Positron-defect profiling in Cd<Subscript>1−x</Subscript>Zn<Subscript>x</Subscript>Te wafers after saw cutting

Near-surface damage induced by saw cutting of ingots of Cd_1−xZn_xTe was investigated by positron-defect depth profiling. The damage extends to several micrometers depth and depends on the cutting apparatus. The samples were polished and etched repeatedly, followed each time by positron-depth profiling. New subsurface damage created during the polishing process is observed. No new damage is observed after etching. Positron-depth profiling is suggested as a diagnostic tool to monitor the quality of sample surfaces.     

3D Topology Preserving Flows for Viewpoint-Based Cortical Unfolding

We present a variational method for unfolding of the cortex based on a user-chosen point of view as an alternative to more traditional global flattening methods, which incur more distortion around the region of interest. Our approach involves three novel contributions. The first is an energy function and its corresponding gradient flow to measure the average visibility of a region of interest of a surface with respect to a given viewpoint. The second is an additional energy function and flow designed to preserve the 3D topology of the evolving surface. The third is a method that dramatically improves the computational speed of the 3D topology preservation approach by creating a tree structure of the 3D surface and using a recursion technique. Experiments results show that the proposed approach can successfully unfold highly convoluted surfaces such as the cortex while preserving their topology during the evolution.     

Secretoneurin and chemoattractant receptor interactions

Regional alterations in cerebral blood flow (CBF) are widely used as a surrogate for neuronal function based on an intact coupling between changes in regional CBF and metabolism, activation-flow coupling (AFC). To further investigate parameters affecting AFC, we have implemented a rat model with electrical forepaw stimulation under alpha-chloralose anesthesia using laser Doppler (LD) measurements of flow parameters through thinned skull over contralateral somatosensory cortex. Signal averaging of the LD response was used to improve reproducibility. A characteristic flow response to electrical forepaw stimulation was reliably recorded from the somatosensory cortex using signal averaging. Stimulation at 5 Hz maximized the LD response, and constant current stimulation up to 1 mA did not induce changes in systemic blood pressure. The shape of the flow response consisted of an initial peak followed by a steady state plateau phase which was observed for stimulation durations longer than 4 s. When individual LD parameters of velocity, red blood cell concentration (CRBC), and cerebral blood flow (CBF) were compared, changes in LDCBF were primarily attributable to changes in LDvelocity rather than LDCRBC. This finding was also observed during hypercapnia. Characterization of AFC in the model provides a background for future studies of the effects of pharmacological manipulation or pathophysiological states.