Operating system noise has been shown to be a key limiter of application scalability in high-end systems. While several studies have attempted to quantify the sources and effects of system interference using user-level mechanisms, there are few published studies on the effect of different kinds of kernel-generated noise on application performance at scale. In this paper, we examine the sensitivity of real-world, large-scale applications to a range of OS noise patterns using a kernel-based noise injection mechanism implemented in the Catamount lightweight kernel. Our results demonstrate the importance of how noise is generated, in terms of frequency and duration, and how this impact changes with application scale. For example, our results show that 2.5% net processor noise at 10,000 nodes can have no impact or can result in over a factor of 20 slowdown for the same application, depending solely on how the noise is generated. We also discuss how the characteristics of the applications we studied, for example computation/communication ratios, collective communication sizes, and other characteristics, related to their tendency to amplify or absorb noise. Finally, we discuss the implications of our findings on the design of new operating systems, middleware, and other system services for high-end parallel systems.

No abstract

Fatigue crack propagation tests with single tensile peak overloads have been performed in 6082-T6 aluminium alloy at several baseline DK levels and stress ratios of 0.05 and 0.25. The tests were carried out at constant DK conditions. Crack closure was monitored in all tests by the compliance technique using a pin microgauge. The observed transient post-overload behaviour is discussed in terms of overload ratio, baseline DK level and stress ratio. The crack closure parameter U was obtained and compared with the crack growth transients. Experimental support is given for the hypothesis that plasticity-induced closure is the main cause of overload retardation for plane stress conditions. Predictions based on crack closure measurements show good correlation with the observed crack growth rates for all the post-overload transients when discontinuous closure is properly taken into account.

Background: Surgical mortality data are collected routinely in high-income countries, yet virtually no low-or middle-income countries have outcome surveillance in place. The aim was prospectively to collect worldwide mortality data following emergency abdominal surgery, comparing findings across countries with a low, middle or high Human Development Index (HDI).Methods: This was a prospective, multicentre, cohort study. Self-selected hospitals performing emergency surgery submitted prespecified data for consecutive patients from at least one 2-week interval during July to December 2014. Postoperative mortality was analysed by hierarchical multivariable logistic regression.

A leaf spot disease of maize occurring in Brazil in the 1980s was described as being caused by the ascomycete Phaeosphaeria maydis (P. Henn) Rane. Payak and Renfro (imperfect form Phyllosticta sp.). Disease symptoms were dark‐green water‐soaked spots that later became necrotic lesions. There are no reports at present in the literature of re‐infection by the fungus under controlled conditions, casting doubt on the true identity of the pathogen. In this study, cytological analyses of lesions at the initial stages did not detect the presence of fungal structures. Bacterial colonies with yellow pigmentation were isolated from the lesions, which reacted positively in hypersensitivity tests in tobacco plants. Maize plants were inoculated with the isolated bacteria. After 72 h incubation in a dew chamber, plants were transferred to a greenhouse, where they remained until evaluation. Typical symptoms of the disease were observed 5–7 days after inoculation of plants, only on treatments inoculated with the bacteria. The bacterium was re‐isolated, which suggests its involvement in the initial phases of disease. The bacterium was identified as Pantoea ananas (synonym Erwinia ananas).

Tuberculosis treatment remains a challenge due to the need to consider, when approaching it, the context of individual and collective health. In addition, social and economic issues have been shown to be variables that need to be considered when it comes to treatment effectiveness. We conducted a critical review of the national and international literature on the treatment of tuberculosis in recent years with the aims of presenting health care workers with recommendations based on the situation in Brazil and better informing decision-making regarding tuberculosis patients so as to minimize morbidity and interrupt disease transmission.

This study reports an experimental investigation of fatigue crack propagation in AlMgSi1‐T6 aluminium alloy using both constant and variable load amplitudes. Crack closure was monitored in all tests by the compliance technique using a pin microgauge. For the constant amplitude tests four different stress ratios were analysed. The crack closure parameter U was calculated and related with ΔK and the stress ratio, R. The threshold of the stress intensity factor range, ΔKth , was also obtained. Fatigue crack propagation tests with single tensile peak overloads have been performed at constant load amplitude conditions. The observed transient post overload behaviour is discussed in terms of the overload ratio, ΔK baseline level and R. The crack closure parameter U trends are compared with the crack growth transients. Experimental support is given for the hypothesis that crack closure is the main factor determining the transient crack growth behaviour following overloads on AlMgSi1‐T6 alloy for plane stress conditions.

In this study low-cycle fatigue tests were performed in two AlMgSi aluminium alloys with different chemical composition, namely 6082-T6 and 6060-T6 alloys, using standard round specimens and tube specimens, respectively. The tests were undertaken in strain control with a strain ratio R e ¼ À1. The cyclic stress-strain curves were determined using one specimen for each imposed strain level. The low-cycle fatigue results are used for the characterisation of the cyclic plastic response and the fatigue live of the alloys. Moreover, the geometry of the hysteresis loops and the occurrence of Masing behaviour are also analysed. The observed behaviour is discussed in terms of the chemical composition of the alloys (Mg 2 Si hardening particles and Mn dispersoid content) and fracture mechanisms. Alloy 6060-T6 exhibits nearly ideal Masing behaviour, while alloy 6082-T6 presents significant deviations from the Masing model. The type of cyclic deformation behaviour in AlMgSi alloys seems to be influenced by the dispersoid phase.