TProf: An energy profiler for task-parallel programs

We present TProf, an energy profiling tool for OpenMP-like task-parallel programs. To compute the energy consumed by each task in a parallel application, TProf dynamically traces the parallel execution and uses a novel technique to estimate the per-task energy consumption. To achieve this estimation, TProf apportions the total processor energy among cores and overcomes the limitation of current works which would otherwise make parallel accounting impossible to achieve. We demonstrate the value of TProf by characterizing a set of task parallel programs, where we find that data locality, memory access patterns and task working sets are responsible for significant variance in energy consumption between seemingly homogeneous tasks. In addition, we identify opportunities for fine-grain energy optimization by applying per-task Dynamic Voltage and Frequency Scaling (DVFS).

Adding metadata to Orc to support reasoning about grid programs

Following earlier work demonstrating the utility of Orc as a means of specifying and reasoning about grid applications we propose the enhancement of such specifications with metadata that provide a means to extend an Orc specification with implementation oriented information. We argue that such specifications provide a useful refinement step in allowing reasoning about implementation related issues ahead of actual implementation or even prototyping. As examples, we demonstrate how such extended specifications can be used for investigating security related issues and for evaluating the cost of handling grid resource faults. The approach emphasises a semi-formal style of reasoning that makes maximum use of programmer domain knowledge and experience.

Cytotoxic assessment of the regulated, co-existing mycotoxins aflatoxin B1, fumonisin B1 and ochratoxin, in single, binary and tertiary mixtures

Aflatoxin B1 (AFB1), ochratoxin A (OTA) and fumonisin B1 (FB1) are contaminants which have been shown to regularly co-occur in a range of foods. However, only a small number of studies have evaluated the interactive effect of binary and tertiary mycotoxins. The present study evaluated the effects of low levels of each mycotoxin in combination at their EU regulatory limits. Toxic effect with respect to cell viability was measured by MTT and neutral red assays, assessing mitochondria and lysosome integrities respectively. Individual toxicity showed that OTA (10 μg/ml) was the most cytotoxic mycotoxin in all three cell lines studied (caco-2, MDBK and raw 264.7). Binary combinations were cytotoxic to the MDBK cell line in the order [OTA/FB1] > [AFB1/FB1] > [AFB1/OTA], whilst all effects observed were classified as being additive. Tertiary combinations of AFB1, FB1 and OTA at the EU regulatory limits were tested and not found to exhibit measurable cytotoxicity in MDBK, caco-2 or raw 264.7 cells. However by increasing these concentrations above the legal limits to OTA (3 μg/ml), FB1 (8 μg/ml) and AFB1 (1.28 μg/ml), cytotoxicity was observed with up to 26% reduction in cell viability and synergistic effects were evident with regard to mitochondrial integrity. © 2014 Elsevier Ltd. All rights reserved.

Cost-Directed Refactoring for Parallel Erlang Programs

This paper presents a new programming methodology for introducing and tuning parallelism in Erlang programs, using source-level code refactoring from sequential source programs to parallel programs written using our skeleton library, Skel. High-level cost models allow us to predict with reasonable accuracy the parallel performance of the refactored program, enabling programmers to make informed decisions about which refactorings to apply. Using our approach, we demonstrate easily obtainable, significant and scalable speedups of up to 21 on a 24-core machine over the sequential code.

Sequence Effects in the Valuation of Multiple Environmental Programs Using the Contingent Valuation Method

In contingent valuation, the willingness to pay for hypothetical programs may be affected by the order in which programs are presented to respondents. With inclusive lists, economic theory suggests that sequence effects should be expected. However, when policy makers allocate public budgets to several environmental programs, they may be interested in assessing the value of the programs without the valuations being affected by the order in which the programs are presented. Using single-bounded dichotomous choice contingent valuation questions, we show that if respondents have the possibility to revise their willingness-to-pay answers, sequence effects are mitigated. (JEL Q51, Q54)