Model-Centric, Context-Aware Software Adaptation

Software must be constantly adapted to changing requirements. The time scale, abstraction level and granularity of adaptations may vary from short-term, fine-grained adaptation to long-term, coarse-grained evolution. Fine-grained, dynamic and context-dependent adaptations can be particularly difficult to realize in long-lived, large-scale software systems. We argue that, in order to effectively and efficiently deploy such changes, adaptive applications must be built on an infrastructure that is not just model-driven, but is both model-centric and context-aware. Specifically, this means that high-level, causally-connected models of the application and the software infrastructure itself should be available at run-time, and that changes may need to be scoped to the run-time execution context. We first review the dimensions of software adaptation and evolution, and then we show how model-centric design can address the adaptation needs of a variety of applications that span these dimensions. We demonstrate through concrete examples how model-centric and context-aware designs work at the level of application interface, programming language and runtime. We then propose a research agenda for a model-centric development environment that supports dynamic software adaptation and evolution.

Collision Detection: Broad Phase Adaptation from Multi-Core to Multi-GPU Architecture

We present in this paper several contributions on the collision detection optimization centered on hardware performance. We focus on the broad phase which is the first step of the collision detection process and propose three new ways of parallelization of the well-known Sweep and Prune algorithm. We first developed a multi-core model takes into account the number of available cores. Multi-core architecture enables us to distribute geometric computations with use of multi-threading. Critical writing section and threads idling have been minimized by introducing new data structures for each thread. Programming with directives, like OpenMP, appears to be a good compromise for code portability. We then proposed a new GPU-based algorithm also based on the "Sweep and Prune" that has been adapted to multi-GPU architectures. Our technique is based on a spatial subdivision method used to distribute computations among GPUs. Results show that significant speed-up can be obtained by passing from 1 to 4 GPUs in a large-scale environment.

Experimental adaptation of wild-type canine distemper virus (CDV) to the human entry receptor CD150.

Canine distemper virus (CDV), a close relative of measles virus (MV), is widespread and well known for its broad host range. When the goal of measles eradication may be achieved, and when measles vaccination will be stopped, CDV might eventually cross the species barrier to humans and emerge as a new human pathogen. In order to get an impression how fast such alterations may occur, we characterized required adaptive mutations to the human entry receptors CD150 (SLAM) and nectin-4 as first step to infect human target cells. Recombinant wild-type CDV-A75/17(red) adapted quickly to growth in human H358 epithelial cells expressing human nectin-4. Sequencing of the viral attachment proteins (hemagglutinin, H, and fusion protein, F) genes revealed that no adaptive alteration was required to utilize human nectin-4. In contrast, the virus replicated only to low titres (10(2) pfu/ml) in Vero cells expressing human CD150 (Vero-hSLAM). After three passages using these cells virus was adapted to human CD150 and replicated to high titres (10(5) pfu/ml). Sequence analyses revealed that only one amino acid exchange in the H-protein at position 540 Asp→Gly (D540G) was required for functional adaptation to human CD150. Structural modelling suggests that the adaptive mutation D540G in H reflects the sequence alteration from canine to human CD150 at position 70 and 71 from Pro to Leu (P70L) and Gly to Glu (G71E), and compensates for the gain of a negative charge in the human CD150 molecule. Using this model system our data indicate that only a minimal alteration, in this case one adaptive mutation, is required for adaptation of CDV to the human entry receptors, and help to understand the molecular basis why this adaptive mutation occurs.

Sweet Puppies and Cute Babies: Perceptual Adaptation to Babyfacedness Transfers across Species

Infant faces are very salient stimuli. The Kindchenschema describes specific features that characterize a cute infant face. In this study we used a visual adaptation paradigm to investigate the universality of the perceptual properties of the Kindchenschema. In Experiment 1, twenty-four participants adapted to cute and less cute human infant faces and in Experiment 2, twenty-four new participants adapted to cute and less cute faces of puppy dogs. In both experiments the task was to assess the cuteness of subsequently presented human infant faces. The results revealed cuteness after-effects for human infant faces in both adaptation conditions, suggesting a common mechanism coding cuteness in human and non-human faces. This study provides experimental evidence for the universality of the well-described concept of the Kindchenschema.

Different adaptation of IGF-I and its IGFBPs in dairy cows during a negative energy balance in early lactation and a negative energy balance induced by feed restriction in mid-lactation

Control of metabolic pathways is a major task of the somatotropic axis and its constituents. Insulinlike growth-factor binding proteins (IGFBPs) bind IGF-I and -II and act as carriers and regulators of their activities in blood, body fluids and tissues. Over two periods of physiological adaptation, this study investigated the binding pattern of IGF-I to IGFBPs in the plasma of 50 multiparous Holstein dairy cows and identified relationships with the hepatic mRNA abundance of IGFBPs and plasma IGF-I during the lactational negative energy balance (NEB) and during a deliberately induced NEB by feed restriction. Period 1 lasted from week 3 antepartum (a.p.) to week 12 postpartum (p.p.) and period 2, the period of feed restriction, started at around 100 DIM and lasted for three weeks with a control (C) and a restricted group (R). Blood samples and liver biopsies were collected in week 3 a.p., and in weeks 1 and 4 p.p. of period 1 and in weeks 0 and 3 of period 2. For column chromatography of IGFBPs, plasma samples of all animals were pooled by group and time points of sampling. Plasma IGF-I dropped from week 3 a.p. to week 1 p.p. and thereafter increased until week 0 (period 2) and did not change up to week 3 of period 2. The binding of IGF-I to plasma IGFBP-1 and -2 increased in period 1 from week 3 a.p. to week 4 p.p., while at the same time it decreased for IGFBP-3. During period 2, the binding of IGF-I to plasma IGFBP-1 and -2 decreased for both groups, but less for R cows. In C cows, the IGF-I binding to IGFBP-3 in plasma increased from week 0 to week 3 of period 2, whereas R cows showed a slight decrease. In period 1, hepatic mRNA abundance of IGFBP-3 followed the plasma IGFBP-3 binding in contrast to the mRNA abundances of IGFBP-1 and -2. The latter increased from week 3 a.p. to week 1 p.p. and decreased afterwards whereas IGF-I binding to IGFBP-1 and -2 increased. In week 3 of period 2, the binding of IGF-I to IGFBP-1 and -2 and their hepatic mRNA abundance were higher in R cows compared to C cows. Hepatic mRNA abundance of IGF-I was consistently positively correlated with plasma IGF-I, especially pronounced during the NEBs in week 1 p.p. (period 1) and in week 3 (period 2) in R cows. While no distinct relation between mRNA abundance of IGFBP-1 and plasma IGF-I was evident, the mRNA abundance of IGFBP-2 was inversely related to plasma IGF-I over all experimental time points independent of treatment. The mRNA abundance of IGFBP-3 was particularly correlated with plasma IGF-I during the 2 experimental stages of a NEB. Obviously IGFBP-3, but not IGFBP-1 and -2, binding in plasma closely followed the respective pattern of hepatic mRNA abundance during the entire experimental period. The fact that changes in the different plasma IGFBPs during altering metabolic stages in different stages of lactation do not always strictly follow their mRNA abundance in liver suggests tissues other than the liver flexibly contributing to the IGFBP pool in plasma as well as a partially post-transcriptional regulation of IGFBP synthesis.

Combining analytical frameworks to assess livelihood vulnerability to climate change and analyse adaptation options

Experts working on behalf of international development organisations need better tools to assist land managers in developing countriesmaintain their livelihoods, as climate change puts pressure on the ecosystemservices that they depend upon. However, current understanding of livelihood vulnerability to climate change is based on a fractured and disparate set of theories andmethods. This reviewtherefore combines theoretical insights from sustainable livelihoods analysis with other analytical frameworks (including the ecosystem services framework, diffusion theory, social learning, adaptive management and transitions management) to assess the vulnerability of rural livelihoods to climate change. This integrated analytical framework helps diagnose vulnerability to climate change,whilst identifying and comparing adaptation options that could reduce vulnerability, following four broad steps: i) determine likely level of exposure to climate change, and how climate change might interact with existing stresses and other future drivers of change; ii) determine the sensitivity of stocks of capital assets and flows of ecosystem services to climate change; iii) identify factors influencing decisions to develop and/or adopt different adaptation strategies, based on innovation or the use/substitution of existing assets; and iv) identify and evaluate potential trade-offs between adaptation options. The paper concludes by identifying interdisciplinary research needs for assessing the vulnerability of livelihoods to climate change.

Adaptation options under climate change for multifunctional agriculture: a simulation study for western Switzerland

Besides its primary role in producing food and fiber, agriculture also has relevant effects on several other functions, such as management of renewable natural resources. Climate change (CC) may lead to new trade-offs between agricultural functions or aggravate existing ones, but suitable agricultural management may maintain or even improve the ability of agroecosystems to supply these functions. Hence, it is necessary to identify relevant drivers (e.g., cropping practices, local conditions) and their interactions, and how they affect agricultural functions in a changing climate. The goal of this study was to use a modeling framework to analyze the sensitivity of indicators of three important agricultural functions, namely crop yield (food and fiber production function), soil erosion (soil conservation function), and nutrient leaching (clean water provision function), to a wide range of agricultural practices for current and future climate conditions. In a two-step approach, cropping practices that explain high proportions of variance of the different indicators were first identified by an analysis of variance-based sensitivity analysis. Then, most suitable combinations of practices to achieve best performance with respect to each indicator were extracted, and trade-offs were analyzed. The procedure was applied to a region in western Switzerland, considering two different soil types to test the importance of local environmental constraints. Results show that the sensitivity of crop yield and soil erosion due to management is high, while nutrient leaching mostly depends on soil type. We found that the influence of most agricultural practices does not change significantly with CC; only irrigation becomes more relevant as a consequence of decreasing summer rainfall. Trade-offs were identified when focusing on best performances of each indicator separately, and these were amplified under CC. For adaptation to CC in the selected study region, conservation soil management and the use of cropped grasslands appear to be the most suitable options to avoid trade-offs.