Distributed control parallelism in multidisciplinary aircraft design

Multidisciplinary design optimization (MDO) for large-scale engineering problems poses many challenges (e.g. the design of an efficient concurrent paradigm for global optimization based on disciplinary analyses, expensive computations over vast data sets, etc.). This work focuses on the application of distributed schemes for massively parallel architectures to MDO problems, as a tool for reducing computation time and solving larger problems. The specific problem considered here is configuration optimization of a high speed civil transport (HSCT), and the efficient parallelization of the embedded paradigm for reasonable design space identification. Two distributed dynamic load balancing techniques (random polling and global round robin with message combining) and two necessary termination detection schemes (global task count and token passing) were implemented and evaluated in terms of effectiveness and scalability to large problem sizes and a thousand processors. The effect of certain parameters on execution time was also inspected. Empirical results demonstrated stable performance and effectiveness for all schemes, and the parametric study showed that the selected algorithmic parameters have a negligible effect on performance. Copyright © 1999 John Wiley & Sons, Ltd.     

A Domain-Specific Language for Generating Dataflow Analyzers

Dataflow analysis is a well-understood and very powerful technique for analyzing programs as part of the compilation process. Virtually all compilers use some sort of dataflow analysis as part of their optimization phase. However, despite being well-understood theoretically, such analyses are often difficult to code, making it difficult to quickly experiment with variants.To address this, we developed a domain-specific language, Analyzer Generator (AG), that synthesizes dataflow analysis phases for Microsoft's Phoenix compiler framework. AG hides the fussy details needed to make analyses modular, yet generates code that is as efficient as the hand-coded equivalent. One key construct we introduce allows IR object classes to be extended without recompiling.Experimental results on three analyses show that AG code can be one-tenth the size of the equivalent handwritten C++ code with no loss of performance. It is our hope that AG will make developing new dataflow analyses much easier.     

Parallel path consistency

Filtering algorithms are well accepted as a means of speeding up the solution of the consistent labeling problem (CLP). Despite the fact that path consistency does a better job of filtering than arc consistency, AC is still the preferred technique because it has a much lower time complexity. We are implementing parallel path consistency algorithms on multiprocessors and comparing their performance to the best sequential and parallel arc consistency algorithms.^(1,2) (See also work by Kerethoet al. ⁽³⁾ and Kasif⁽⁴⁾) Preliminary work has shown linear performance increases for parallelized path consistency and also shown that in many cases performance is significantly better than the theoretical worst case. These two results lead us to believe that parallel path consistency may be a superior filtering technique. Finally, we have implemented path consistency as an outer product computation and have obtained good results (e.g., linear speedup on a 64K-node Connection Machine 2).     

Improving inter-destination synchronization in hierarchical reliable multicast

In hierarchical reliable multicast schemes, the number of repair proxies and their locations influence inter-destination synchronization. The inter-destination synchronization in multicast environments means the adjustment of the output timing among destinations over the Internet. Improving the inter-destination synchronization is beneficial to collaborative applications such as multi-conference system and multi-playing online game. In this paper, we propose a scheme to find the optimal locations of repair proxies that can improve inter-destination synchronization maximally in heterogeneous network environments. The simulation results show that if repair proxies are placed by the proposed scheme, delivery delay fairness of inter-destination can be improved by 0.05 maximally. In addition, we perform t-Test on the simulation results in order to verify that our optimal placement improves synchronization.     

Circadian variation in the onset of acute pulmonary edema and associated acute myocardial infarction in diabetic and nondiabetic patients

It is known that most acute cardiovascular events exhibit a circadian rhythm in their onset. The authors describe differences in the circadian rhythm of onset of acute pulmonary edema and associated acute myocardial infarction in diabetic and nondiabetic patients.     

A pilot-plant study of the adsorptive micellar flocculation process: Optimum design and operation

A pilot-plant scale study of the adsorptive micellar flocculation (AMF) process is presented for the first time, and builds on the solid foundation provided by previous fundamental laboratory studies. A number of experimental tests were conducted to validate the feasibility of using an AMF pilot rig to remove phenol from aqueous effluent streams. Several key factors, including flocculation time, floc settling time, optimum air flow-rate for agitation, and flocculant dosage, were determined. Residual concentrations in treated water indicated a surfactant removal efficiency of 95–98%, and pollutant removal reached 78% in two stages of process operation. A strategy for the separate recycling of surfactant, flocculant and pollutant has already been developed. The present study has thus made further progress in developing, testing, validating and optimising the operating conditions of a pilot-scale process. As such, it has demonstrated the feasibility of scaling up from the bench-scale to a commercial, continuously operated unit.     

Thermosolutal convection during directional solidification

During solidification of a binary alloy at constant velocity vertically upward, thermosolutal convection can occur if the solute rejected at the crystal-melt interface decreases the density of the melt. We assume that the crystal-melt interface remains planar and that the flow field is periodic in the horizontal direction. The time-dependent nonlinear differential equations for fluid flow, concentration, and temperature are solved numerically in two spatial dimensions for small Prandtl numbers and moderately large Schmidt numbers. For slow solidification velocities, the thermal field has an important stabilizing influence: near the onset of instability the flow is confined to the vicinity of the crystal-melt interface. Further, for slow velocities, as the concentration increases, the horizontal wavelength of the flow decreases rapidly — a phenomenon also indicated by linear stability analysis. The lateral in-homogeneity in solute concentration due to convection is obtained from the calculations. For a narrow range of solutal Rayleigh numbers and wavelengths, the flow is periodic in time. Formerly with the Mathematical Analysis Division, Center for Applied Mathematics, National Bureau of Standards, Washington, DC 20234. This paper is based on a presentation made at the symposium “Fluid Flow at Solid-Liquid Interfaces” held at the fall meeting of the TMS-AIME in Philadelphia, PA on October 5, 1983 under the TMS-AIME Solidification Committee.     

Quality changes of salmon (Oncorhynchus gorbuscha) muscle during thermal processing

ABSTRACT:  The objective of this study was to investigate quality changes of salmon fillet muscle during thermal sterilization processes. Small samples (D 30 mm × H 6 mm) from the central dorsal region were heated in an oil bath at 121.1 °C for periods varying from 5 to 120 min. The quality variations along the longitudinal axis of salmon fillets (raw and heated) were examined. The quality properties studied included shear force, color, cook loss, and shrinkage. To minimize the influence of the heterogeneity of the salmon muscle, a multiple thin blade texture device was developed for shear force measurement and a computer vision system was used to facilitate accurate measurements of color and shrinkage. The red muscle was firmer than the white muscle in the raw but not in heated samples. Muscle from the central dorsal region had a lower cook loss and less shrinkage than samples from either the anterior or posterior region following heating. The greatest change in quality occurred within the 1st 10 min of heating at 121.1 °C. Shear force measurements following heating indicated 2 peaks, one corresponding to 5 min and the second for 60 min processing at 121.1 °C. Possible mechanisms were discussed.     

BAX regulates follicular endowment in mice

It is believed that the endowment of primordial follicles in mammalian ovaries is finite. Once follicles are depleted, infertility ensues. Thus, the size of the initial endowment has consequences for fertility and reproductive longevity. Follicular endowment is comprised of various processes that culminate with the incorporation of meiosis-arrested oocytes into primordial follicles. Apoptosis is prominent during follicular endowment, and apoptosis regulatory genes are involved in its regulation. Conflicting data exist with regard to the role of the proapoptotic Bcl-2 associated X protein (BAX) in follicular endowment. Therefore, we investigated the role of BAX during follicular endowment in embryonic and neonatal ovaries. We found that BAX is involved in regulating follicular endowment in mice. Deletion of Bax yields increased oocyte numbers in embryonic ovaries and increased follicle numbers in neonatal ovaries when compared with wild-type ovaries. Increased follicular endowment in Bax -/- ovaries is not due to enhanced germ cell viability. Further, it is not due to an increased primordial germ cell (PGC) allotment, a delay in the onset of meiosis, or altered proliferative activity of oogonia. Instead, our data suggest that the regulatory activity of BAX in follicular endowment likely occurs during PGC migration, prior to PGC colonization of the gonad.     

A process performance index and its application to optimization of the rtm process

Resin transfer molding (RTM) is a promising manufacturing process for hig formance composite materials. However, the fact that RTM process design has traditionally been an expensive, time‐consuming trial‐and‐error procedure has p ited its wide application base. This paper proposes a solution to that problem—a simulation‐based optimum process design scheme for RTM. This scheme ei engineers to determine the optimum locations of injection gates and vents so both process efficiency and high part quality can be ensured. Essential to this mum process design scheme is a process performance index, which is defined respect to the major factors influencing RTM process efficiency and part quality This index is then used as the objective function for the RTM process design optimization model. Gate and vent locations are the process design parameters optimized. All data is obtained by running an RTM simulation program, and th netic algorithm (GA) is employed to carry out the optimization procedure for design parameters. It is found that constant pressure optimization will yi process with a short flow path, whereas constant flow optimization will yield process with smooth and vent‐oriented flow pattern. Although there is no dry factor in the objective function, it is interesting to note that both constant pres and constant flow optimization procedures result in process designs with a mil mum probability of dry spot formation. This study finds that, in general, cons flow optimization should be employed if injection pressure is not a major cone otherwise, constant pressure optimization should be used. Two case studies presented to illustrate the efficacy of this approach.