Analysis of multi-cluster computing systems with processor heterogeneity

This paper addresses the problem of performance modeling of heterogeneous multi-cluster computing systems. We present an analytical model that can be employed to explore the effectiveness of different design approaches so that one can have an intelligent choice during design and evaluation of a cost effective large-scale heterogeneous distributed computing system. The proposed model considers stochastic quantities as well as processor heterogeneity of the target system. The analysis is based on a parametric fat-tree network, the m-port n-tree, and a deterministic routing algorithm. The correctness of the proposed model is validated through comprehensive simulation of different types of clusters.

Analytical modeling of communication latency in multi-cluster systems

This paper addresses the problem of performance modeling of large-scale distributed systems with emphasis on communication networks in heterogeneous multi-cluster systems. The study of interconnection networks is important because the overall performance of a distributed system is often critically hinged on the effectiveness of this part. We present an analytical model to predict message latency in multi-cluster systems in the presence of processor heterogeneity. The model is validated through comprehensive simulation, which demonstrates that the proposed model exhibits a good degree of accuracy for various system sizes and under different operating conditions.

Guest editorial: foreword to the special issue on aggregation operators

The fourteen papers in this special section are devoted to aggregation operators with respect to knowledge based systems.

Fitting aggregation operators to data

Theoretical advances in modelling aggregation of information produced a wide range of aggregation operators, applicable to almost every practical problem. The most important classes of aggregation operators include triangular norms, uninorms, generalised means and OWA operators.
With such a variety, an important practical problem has emerged: how to fit the parameters/ weights of these families of aggregation operators to observed data? How to estimate quantitatively whether a given class of operators is suitable as a model in a given practical setting? Aggregation operators are rather special classes of functions, and thus they require specialised regression techniques, which would enforce important theoretical properties, like commutativity or associativity. My presentation will address this issue in detail, and will discuss various regression methods applicable specifically to t-norms, uninorms and generalised means. I will also demonstrate software implementing these regression techniques, which would allow practitioners to paste their data and obtain optimal parameters of the chosen family of operators.

Pointwise construction of Lippschitz aggregation operators with specific properties

This paper describes an approach to pointwise construction of general aggregation operators, based on monotone Lipschitz approximation. The aggregation operators are constructed from a set of desired values at certain points, or from empirically collected data. It establishes tight upper and lower bounds on Lipschitz aggregation operators with a number of different properties, as well as the optimal aggregation operator, consistent with the given values. We consider conjunctive, disjunctive and idempotent n-ary aggregation operators; p-stable aggregation operators; various choices of the neutral element and annihilator; diagonal, opposite diagonal and marginal sections; bipolar and double aggregation operators. In all cases we provide either explicit formulas or deterministic numerical procedures to determine the bounds. The findings of this paper are useful for construction of aggregation operators with specified properties, especially using interpolation schemata.

Interpolatory type construction of general aggregation operators

In this chapter we examine a number of methods to construct aggregation operators of interpolatory type for specific applications. The construction is based on the desired values of the aggregation operator at certain prototypical points, and on other desired properties, such as, conjuctive, disjunctive or averaging behaviour, symmetry and marginals.

Identification of weights in aggregation operators

This chapter provides a review of various techniques for identification of weights in generalized mean and ordered weighted averaging aggregation operators, as well as identification of fuzzy measures in Choquet integral based operators. Our main focus is on using empirical data to compute the weights. We present a number of practical algorithms to identify the best aggregation operator that fits the data.

Use of a decision aid for prenatal testing of fetal abnormalities to improve women's informed decision making: a cluster randomised controlled trial [ISRCTN22532458]

Objective To evaluate the effectiveness of a decision aid for prenatal testing of fetal abnormalities compared with a pamphlet in supporting women's decision making.
Design A cluster randomised controlled trial.
Setting Primary health care.
Population Women in early pregnancy consulting a GP.
Methods GPs were randomised to provide women with either a decision aid or a pamphlet. The decision aid was a 24-page booklet designed using the Ottowa Decision Framework. The pamphlet was an existing resource available in the trial setting.
Main outcome measures Validated scales were used to measure the primary outcomes, informed choice and decisional conflict, and the secondary outcomes, anxiety, depression, attitudes to the pregnancy/fetus and acceptability of the resource. Outcomes were measured at 14 weeks of gestation from questionnaires that women completed and returned by post.
Findings Women in the intervention group were more likely to make an informed decision 76% (126/165) than those in the control group 65% (107/165) (adjusted OR 2.08; 95% CI 1.14–3.81). A greater proportion of women in the intervention group 88% (147/167) had a 'good' level of knowledge than those in the control group 72% (123/171) (adjusted OR 3.43; 95% CI 1.79–6.58). Mean (SD) decisional conflict scores were low in both groups, decision aid 1.71 (0.49), pamphlet 1.65 (0.55) (adjusted mean difference 0.10; 95% CI −0.02 to 0.22). There was no strong evidence of differences between the trial arms in the measures of psychological or acceptability outcomes.
Conclusion A tailored prenatal testing decision aid plays an important role in improving women's knowledge of first and second trimester screening tests and assisting them to make decisions about screening and diagnostic tests that are consistent with their values.

The infant feeding activity and nutrition trial (INFANT) an early intervention to prevent childhood obesity : cluster-randomised controlled trial

Background : Multiple factors combine to support a compelling case for interventions that target the development of obesity-promoting behaviours (poor diet, low physical activity and high sedentary behaviour) from their inception. These factors include the rapidly increasing prevalence of fatness throughout childhood, the instigation of obesity-promoting behaviours in infancy, and the tracking of these behaviours from childhood through to adolescence and adulthood. The Infant Feeding Activity and Nutrition Trial (INFANT) aims to determine the effectiveness of an early childhood obesity prevention intervention delivered to first-time parents. The intervention, conducted with parents over the infant's first 18 months of life, will use existing social networks (first-time parent's groups) and an anticipatory guidance framework focusing on parenting skills which support the development of positive diet and physical activity behaviours, and reduced sedentary behaviours in infancy.

Methods/Design : This cluster-randomised controlled trial, with first-time parent groups as the unit of randomisation, will be conducted with a sample of 600 first-time parents and their newborn children who attend the first-time parents' group at Maternal and Child Health Centres. Using a two-stage sampling process, local government areas in Victoria, Australia will be randomly selected at the first stage. At the second stage, a proportional sample of first-time parent groups within selected local government areas will be randomly selected and invited to participate. Informed consent will be obtained and groups will then be randomly allocated to the intervention or control group.

Discussion : The early years hold promise as a time in which obesity prevention may be most effective. To our knowledge this will be the first randomised trial internationally to demonstrate whether an early health promotion program delivered to first-time parents in their existing social groups promotes healthy eating, physical activity and reduced sedentary behaviours. If proven to be effective, INFANT may protect children from the development of obesity and its associated social and economic costs.

Load balancing based concurrent execution of NAS parallel benchmarks with BYTE sequential benchmarks on a non-dedicated cluster

Dedicated clusters are becoming commonly used for high performance parallel processing. Computers of a non-dedicated cluster are often idle or lightly loaded. These under utilised computers can be employed to execute parallel applications. Thus, they have to be shared by parallel and sequential applications, which could lead to the improvement of their execution performance. There is a lack of experimental study showing the behaviour and performance of executing parallel and sequential applications concurrently on a non-dedicated cluster. We present the result of an experimental study into load balancing of a mixture of parallel and sequential applications on a non-dedicated cluster.

A study of the concurrent execution of parallel and sequential applications on a non-dedicated cluster

Computers of a non-dedicated cluster are often idle (users attend meetings, have lunch or coffee breaks) or lightly loaded (users carry out simple computations to support problem solving activities). These underutilised computers can be employed to execute parallel applications. Thus, these computers can be shared by parallel and sequential applications, which could lead to the improvement of their execution performance. However, there is a lack of experimental study showing the applications’ performance and the system utilization of executing parallel and sequential applications concurrently and concurrent execution of multiple parallel applications on a non-dedicated cluster. Here we present the result of an experimental study into load balancing based scheduling of mixtures of NAS Parallel Benchmarks and BYTE sequential applications on a very low cost non-dedicated cluster. This study showed that the proposed sharing provided performance boost as compared to the execution of the parallel load in isolation on a reduced number of computers and better cluster utilization. The results of this research were used not only to validate other researchers’ result generated by simulation but also to support our research mission of widening the use of non-dedicated clusters. Our promising results obtained could promote further research studies to convince universities, business and industry, which require a large amount of computing resources, to run parallel applications on their already owned non-dedicated clusters.