On generating set index functions for randomized caches

Caches hide the growing latency of accesses to the main memory from the processor by storing the most recently used data on-chip. To limit the search time through the caches, they are organized in a direct mapped or set-associative way. Such an organization introduces many conflict misses that hamper performance. This paper studies randomizing set index functions, a technique to place the data in the cache in such a way that conflict misses are avoided. The performance of such a randomized cache strongly depends on the randomization function. This paper discusses a methodology to generate randomization functions that perform well over a broad range of benchmarks. The methodology uses profiling information to predict the conflict miss rate of randomization functions. Then, using this information, a search algorithm finds the best randomization function. Due to implementation issues, it is preferable to use a randomization function that is extremely simple and can be evaluated in little time. For these reasons, we use randomization functions where each randomized address bit is computed as the XOR of a subset of the original address bits. These functions are chosen such that they operate on as few address bits as possible and have few inputs to each XOR. This paper shows that to index a 2(m)-set cache, it suffices to randomize m+2 or m+3 address bits and to limit the number of inputs to each XOR to 2 bits to obtain the full potential of randomization. Furthermore, it is shown that the randomization function that we generate for one set of benchmarks also works well for an entirely different set of benchmarks. Using the described methodology, it is possible to reduce the implementation cost of randomization functions with only an insignificant loss in conflict reduction.

Highly accurate and efficient evaluation of randomising set index functions

Randomising set index functions can reduce the number of conflict misses in data caches by spreading the cache blocks uniformly over all sets. Typically, the randomisation functions compute the exclusive ors of several address bits. Not all randomising set index functions perform equally well, which calls for the evaluation of many set index functions. This paper discusses and improves a technique that tackles this problem by predicting the miss rate incurred by a randomisation function, based on profiling information. A new way of looking at randomisation functions is used, namely the null space of the randomisation function. The members of the null space describe pairs of cache blocks that are mapped to the same set. This paper presents an analytical model of the error made by the technique and uses this to propose several optimisations to the technique. The technique is then applied to generate a conflict-free randomisation function for the SPEC benchmarks. (C) 2003 Elsevier Science B.V. All rights reserved.

Obesity associated severe asthma represents a distinct clinical phenotype - Analysis of the British Thoracic Society Difficult Asthma Registry patient cohort according to body mass index

BACKGROUND: Obesity has emerged as a risk factor for the development of asthma and it may also influence asthma control and airways inflammation. However, the role of obesity in severe asthma remains unclear. OBJECTIVE: To explore the association between obesity (defined by BMI) and severe asthma. METHODS: Data from the National Registry for dedicated UK Difficult Asthma Services were used to compare patient demographics, disease characteristics and healthcare utilisation between three body mass index (BMI) categories (normal weight: 18.5 -24.99, overweight: 25 -29.99, obese: =30) in a well characterised group of severe asthmatic adults. RESULTS: The study population consisted of 666 severe asthmatics with a median BMI of 29.8 (interquartile range 22.5 -34.0). The obese group exhibited greater asthma medication requirements in terms of maintenance corticosteroid therapy (48.9% versus 40.4% and 34.5% in the overweight and normal weight groups, respectively), steroid burst therapy and short-acting ß2-agonist (SABA) use per day. Significant differences were seen with gastro-oesophageal reflux disease (GORD) (53.9% versus 48.1% and 39.7% in the overweight and normal weight groups, respectively) and proton pump inhibitor (PPI) use. Bone density scores were higher in the obese group, whilst pulmonary function testing revealed a reduced FVC and raised Kco. Serum IgE levels decreased with increasing BMI and the obese group were more likely to report eczema, but less likely to have a history of nasal polyps. CONCLUSIONS: Severe asthmatics display particular characteristics according to BMI that support the view that obesity associated severe asthma may represent a distinct clinical phenotype.1Royal Brompton Hospital, London, UK;2Department of Computing, Imperial College, UK3Airways Disease, National Heart & Lung Institute, Imperial College, UK;4Centre for infection and immunity, Queen's University of Belfast, UK;5University of Leicester, UK;6The University of Manchester and University Hospital of South Manchester, UK;7Birmingham Heartlands Hospital, University of Birmingham, UK;8Gartnavel General Hospital, University of Glasgow, UK;9Glasgow Royal Infirmary, Glasgow, UKCorrespondence: Dr Andrew N. Menzies-Gow, Royal Brompton Hospital, Fulham Road, London SW3 6HP.

Optical transmission of periodic annular apertures in metal film on high-refractive index substrate: The role of the nanopillar shape

The influence of annular aperture parameters on the optical transmission through arrays of coaxial apertures in a metal film on high refractive index substrates has been investigated experimentally and numerically. It is shown that the transmission resonances are related to plasmonic crystal effects rather than frequency cutoff behavior associated with annular apertures. The role of deviations from ideal aperture shape occurring during the fabrication process has also been studied. Annular aperture arrays are often considered in many applications for achieving high optical transmission through metal films and understanding of nanofabrication tolerances are important. (C) 2010 American Institute of Physics.

The impact of body mass index on maternal and neonatal outcomes: a retrospective study in a UK obstetric population, 2004-2011

Objective: To assess the prevalence of overweight and obesity, and the impact of body mass index (BMI) on maternal and neonatal outcomes, in a UK obstetric population.
Design: Retrospective study.
Setting: A tertiary referral unit in Northern Ireland.
Population: A total of 30 298 singleton pregnancies over an 8-year period, 2004–2011.
Methods: Women were categorised according to World Health Organization classification: underweight (BMI < 18.50 kg/m2); normal weight (BMI 18.50–24.99 kg/m2; reference group); overweight (BMI 25.00–29.99 kg/m2); obese class I (BMI 30.00–34.99 kg/m2); obese class II (BMI 35–39.99 kg/m2); and obese class III (BMI = 40 kg/m2). Maternal and neonatal outcomes were examined using logistic regression, adjusted for confounding variables.
Main outcome measures: Maternal and neonatal outcomes.
Results: Compared with women of normal weight, women who were overweight or obese class I were at significantly increased risk of hypertensive disorders of pregnancy (OR 1.9, 99% CI 1.7–2.3; OR 3.5, 99% CI 2.9–4.2); gestational diabetes mellitus (OR 1.7, 99% CI 1.3–2.3; OR 3.7, 99% CI 2.8–5.0); induction of labour (OR 1.2, 99% CI 1.1–1.3; OR 1.3, 99% CI 1.2–1.5); caesarean section (OR 1.4, 99% CI 1.3–1.5; OR 1.8, 99% CI 1.6–2.0); postpartum haemorrhage (OR 1.4, 99% CI 1.3–1.5; OR 1.8, 1.6–2.0); and macrosomia (OR 1.5, 99% CI 1.3–1.6; OR 1.9, 99% CI 1.6–2.2), with the risks increasing for obese classes II and III. Women in obese class III were at increased risk of preterm delivery (OR 1.6, 99% CI 1.1–2.5), stillbirth (OR 3.0, 99% CI 1.0–9.3), postnatal stay > 5 days (OR 2.1, 99% CI 1.5–3.1), and infant requiring admission to a neonatal unit (OR 1.6, 99% CI 1.0–2.6).
Conclusions: By categorising women into overweight and obesity subclassifications (classes I –III), this study clearly demonstrates an increasing risk of adverse outcomes across BMI categories, with women who are overweight also at significant risk.
Keywords Body mass index, maternal and neonatal outcomes,obesity, pregnancy

Economic globalization, inequality and body mass index:a cross-national analysis of 127 countries

This article investigates to what extent the worldwide increase in body mass index (BMI) has been affected by economic globalization and inequality. We used time-series and longitudinal cross-national analysis of 127 countries from 1980 to 2008. Data on mean adult BMI were obtained from the Global Burden of Metabolic Risk Factors of Chronic Diseases Collaborating Group. Globalization was measured using the Swiss Economic Institute (KOF) index of economic globalization. Economic inequality between countries was measured with the mean difference in gross domestic product per capita purchasing power parity in international dollars. Economic inequality within countries was measured using the Gini index from the Standardized World Income Inequality Database. Other covariates including poverty, population size, urban population, openness to trade and foreign direct investment were taken from the World Development Indicators (WDI) database. Time-series regression analyses showed that the global increase in BMI is positively associated with both the index of economic globalization and inequality between countries, after adjustment for covariates. Longitudinal panel data analyses showed that the association between economic globalization and BMI is robust after controlling for all covariates and using different estimators. The association between economic inequality within countries and BMI, however, was significant only among high-income nations. More research is needed to study the pathways between economic globalization and BMI. These findings, however, contribute to explaining how contemporary globalization can be reformed to promote better health and control the global obesity epidemic. © 2013 Copyright Taylor and Francis Group, LLC.

Nonparametric Estimation of Data Dimensionality Prior to Data Compression: the case of the Human Development Index

In many applications in applied statistics researchers reduce the complexity of a data set by combining a group of variables into a single measure using factor analysis or an index number. We argue that such compression loses information if the data actually has high dimensionality. We advocate the use of a non-parametric estimator, commonly used in physics (the Takens estimator), to estimate the correlation dimension of the data prior to compression. The advantage of this approach over traditional linear data compression approaches is that the data does not have to be linearized. Applying our ideas to the United Nations Human Development Index we find that the four variables that are used in its construction have dimension three and the index loses information.

Effect of cataract extraction on the Glaucoma Progression Index (GPI) in glaucoma patients

PURPOSE: To determine the effect of cataract extraction on the glaucoma progression index (GPI) in glaucoma patients with coexisting cataract.

PATIENTS AND METHODS: This is a retrospective noncomparative study. Consecutive eligible patients with glaucoma who underwent phacoemulsification alone or in combination with augmented trabeculectomy were included. All patients had Swedish Interactive Threshold Algorithm-standard 24-2 visual fields within 10 months of surgery. Exclusion criteria included other ocular morbidity, intraoperative complications, and perimetric reliability indices greater than 33%. Comparison was made between the immediate visual fields before and after surgery. The main outcome measure was the change in GPI. Changes in the pattern standard deviation (PSD) and mean deviation (MD) were also assessed. Comparison of means was performed with the paired t test.

RESULTS: Thirty-three eyes of 33 patients (all Whites) were analyzed. The mean age at surgery was 77.0+/-8.7 years. Visual field tests were performed 3.3+/-3.0 months SD before surgery and 5.4+/-2.6 months after surgery. There was a statistically significant increase in the GPI after cataract surgery (from 71.5+/-18.5% to 74.6+/-17.1%; P=0.02). The improvement in MD was also statistically significant (from -11.8+/-5.3 to -10.2+/-5.3 dB; P <0.01), but the change in PSD did not reach statistical significance.

CONCLUSIONS: Uncomplicated cataract extraction resulted in a statistically significant improvement in the 24-2 Swedish Interactive Threshold Algorithm-standard GPI and MD, but not in PSD. Both the MD and the GPI may be influenced by lens opacities, which could make detection of glaucoma visual field progression more difficult for clinicians in glaucoma patients with concurrent cataract.

Perimetric progression in open angle glaucoma and the Visual Field Index (VFI)

PURPOSE: To evaluate the changes in the Visual Field Index (VFI) in eyes with perimetric glaucomatous progression, and to compare these against stable glaucoma patients.

PATIENTS AND METHODS: Consecutive patients with open angle glaucoma with a minimum of 6 reliable visual fields and 2 years of follow-up were identified. Perimetric progression was assessed by 4 masked glaucoma experts from different units, and classified into 3 categories: "definite progression," "suspected progression," or "no progression." This was compared with the Glaucoma Progression Analysis (GPA) II and VFI linear regression analysis, where progression was defined as a negative slope with significance of <5%.

RESULTS: Three hundred ninety-seven visual fields from 51 eyes of 39 patients were assessed. The mean number of visual fields was 7.8 (SD 1.1) per eye, and the mean follow-up duration was 63.7 (SD 13.4) months. The mean VFI linear regression slope showed an overall statistically significant difference (P<0.001, analysis of variance) for each category of progression. Using expert consensus opinion as the reference standard, both VFI analysis and GPA II had high specificity (0.93 and 0.90, respectively), but relatively low sensitivity (0.45 and 0.41, respectively).

CONCLUSIONS: The mean VFI regression slope in our cohort of eyes without perimetric progression showed a statistically significant difference compared with those with suspected and definite progression. VFI analysis and GPA II both had similarly high specificity but low sensitivity when compared with expert consensus opinion.