Analysis of Dependence Tracking Algorithms for Task Dataflow Execution

Processor architectures has taken a turn towards many-core processors, which integrate multiple processing cores on a single chip to increase overall performance, and there are no signs that this trend will stop in the near future. Many-core processors are harder to program than multi-core and single-core processors due to the need of writing parallel or concurrent programs with high degrees of parallelism. Moreover, many-cores have to operate in a mode of strong scaling because of memory bandwidth constraints. In strong scaling increasingly finer-grain parallelism must be extracted in order to keep all processing cores busy.

Task dataflow programming models have a high potential to simplify parallel program- ming because they alleviate the programmer from identifying precisely all inter-task de- pendences when writing programs. Instead, the task dataflow runtime system detects and enforces inter-task dependences during execution based on the description of memory each task accesses. The runtime constructs a task dataflow graph that captures all tasks and their dependences. Tasks are scheduled to execute in parallel taking into account dependences specified in the task graph.

Several papers report important overheads for task dataflow systems, which severely limits the scalability and usability of such systems. In this paper we study efficient schemes to manage task graphs and analyze their scalability. We assume a programming model that supports input, output and in/out annotations on task arguments, as well as commutative in/out and reductions. We analyze the structure of task graphs and identify versions and generations as key concepts for efficient management of task graphs. Then, we present three schemes to manage task graphs building on graph representations, hypergraphs and lists. We also consider a fourth edge-less scheme that synchronizes tasks using integers. Analysis using micro-benchmarks shows that the graph representation is not always scalable and that the edge-less scheme introduces least overhead in nearly all situations.

Risk assessment of neonatal excipient exposure:Lessons from food safety and other areas

Newborn babies can require significant amounts of medication containing excipients intended to improve the drug formulation. Most medicines given to neonates have been developed for adults or older children and contain excipients thought to be safe in these age groups. Many excipients have been used widely in neonates without obvious adverse effects. Some excipients may be toxic in high amounts in which case they need careful risk assessment. Alternatively, it is conceivable that ill-founded fears about excipients mean that potentially useful medicines are not made available to newborn babies. Choices about excipient exposure can occur at several stages throughout the lifecycle of a medicine, from product development through to clinical use. Making these choices requires a scalable approach to analysing the overall risk. In this contribution we examine these issues.

Mature students – participation in maths support and progression

Maths support at the Learning Development Service takes the form of drop-in contact, one-to-one appointments and workshops. Analysis over three years from 2010/11 to 2012/13 shows that 45% of one-to-one appointments involved a mature student, defined at Queen’s University Belfast (QUB) as one who has had a break in full-time study (normally a minimum of two years). This is a very high proportion given that 4% of students at QUB are mature. Considering engineering undergraduates only, the fraction of one-to-one appointments involving mature students was also 45%. This study can report a wide variation in terms of progression of mature students in engineering and aims to consider how more traditional undergraduate learners could be persuaded to adopt the attitudes of mature students in partaking of maths support.

Disability benefit growth and disability reform in the US: lessons from other OECD nations

Unsustainable growth in program costs and beneficiaries, together with a growing recognition that even people with severe impairments can work, led to fundamental disability policy reforms in the Netherlands, Sweden, and Great Britain. In Australia, rapid growth in disability recipiency led to more modest reforms. Here we describe the factors driving unsustainable DI program growth in the U.S., show their similarity to the factors that led to unsustainable growth in these other four OECD countries, and discuss the reforms each country implemented to regain control over their cash transfer disability program. Although each country took a unique path to making and implementing fundamental reforms, shared lessons emerge from their experiences.

On the potential of significance-driven execution for energy-aware HPC

Dynamic Voltage and Frequency Scaling (DVFS) exhibits fundamental limitations as a method to reduce energy consumption in computing systems. In the HPC domain, where performance is of highest priority and codes are heavily optimized to minimize idle time, DVFS has limited opportunity to achieve substantial energy savings. This paper explores if operating processors Near the transistor Threshold Volt- age (NTV) is a better alternative to DVFS for break- ing the power wall in HPC. NTV presents challenges, since it compromises both performance and reliability to reduce power consumption. We present a first of its kind study of a significance-driven execution paradigm that selectively uses NTV and algorithmic error tolerance to reduce energy consumption in performance- constrained HPC environments. Using an iterative algorithm as a use case, we present an adaptive execution scheme that switches between near-threshold execution on many cores and above-threshold execution on one core, as the computational significance of iterations in the algorithm evolves over time. Using this scheme on state-of-the-art hardware, we demonstrate energy savings ranging between 35% to 67%, while compromising neither correctness nor performance.

Lessons from 'The Wire': Epistemological Reflections on the Practice of Sociological Research

Recent debates and controversies have highlighted several issues surrounding sociological research, which relate to the general conditions under which it is undertaken and how this is changing. There is a pressing need to respond to these issues as a whole, in particular by examining what they tell us about research practices. This article argues that a consideration of themes raised by the American television drama The Wire is useful for facilitating such a response, since it may be read as a discussion of working conditions within neoliberal societies. The following themes are pertinent here: the need to reflect upon the terms by which research is framed by funders, to take adequate time to conduct and complete research, and to encourage critical debate within research. Whilst these relate to influential epistemological discussions by Pierre Bourdieu and Michael Burawoy, this reading of The Wire is particularly helpful for highlighting the practical and inter-relational situations in which sociological research is carried out but which tend not to receive the systematic attention they deserve.

Optimization by quantum annealing: Lessons from simple cases

We investigate the basic behavior and performance of simulated quantum annealing (QA) in comparison with classical annealing (CA). Three simple one-dimensional case study systems are considered: namely, a parabolic well, a double well, and a curved washboard. The time-dependent Schrodinger evolution in either real or imaginary time describing QA is contrasted with the Fokker-Planck evolution of CA. The asymptotic decrease of excess energy with annealing time is studied in each case, and the reasons for differences are examined and discussed. The Huse-Fisher classical power law of double-well CA is replaced with a different power law in QA. The multiwell washboard problem studied in CA by Shinomoto and Kabashima and leading classically to a logarithmic annealing even in the absence of disorder turns to a power-law behavior when annealed with QA. The crucial role of disorder and localization is briefly discussed.

Dyslexia in higher education: Implications for maths anxiety, statistics anxiety and psychological well-being

This study examined levels of mathematics and statistics anxiety, as well as general mental health amongst undergraduate students with dyslexia (n = 28) and those without dyslexia (n = 71). Students with dyslexia had higher levels of mathematics anxiety relative to those without dyslexia, while statistics anxiety and general mental health were comparable for both reading ability groups. In terms of coping strategies, undergraduates with dyslexia tended to use planning-based strategies and seek instrumental support more frequently than those without dyslexia. Higher mathematics anxiety was associated with having a dyslexia diagnosis, as well as greater levels of worrying, denial, seeking instrumental support and less use of the positive reinterpretation coping strategy. By contrast, statistics anxiety was not predicted by dyslexia diagnosis, but was instead predicted by overall worrying and the use of denial and emotion focused coping strategies. The results suggest that disability practitioners should be aware that university students with dyslexia are at risk of high mathematics anxiety. Additionally, effective anxiety reduction strategies such as positive reframing and thought challenging would form a useful addition to the support package delivered to many students with dyslexia.

Political Extremism in the 1920s and 1930s: Do German Lessons Generalize?

We examine the impact of the Great Depression on the share of votes for right-wing extremists in elections in the 1920s and 1930s. We confirm the existence of a link between political extremism and economic hard times as captured by growth or contraction of the economy. What mattered was not simply growth at the time of the election, but cumulative growth performance. The impact was greatest in countries with relatively short histories of democracy, with electoral systems that created low hurdles to parliamentary representation, and which had been on the losing side in World War I.