Parallelization of WHILE loops on pipelined architectures

Modulo scheduling theory can be applied successfully to overlap Fortran DO loops on pipelined computers issuing multiple operations per cycle both with and without special loop architectural support. This paper shows that a broader class of loops—REPEAT-UNTIL, WHILE, and loops with more than one exit, in which the trip count is not known beforehand—can also be overlapped efficiently on multiple-issue pipelined machines. The approach is described with respect to a specific machine model, but it can be extended to other models. Special features in the architecture, as well as compiler representations for accelerating these loop constructs, are discussed. Performance results are presented for a few select examples.An earlier version of this paper was presented at Supercomputing '90.     

Targeting optimum resource allocation using reverse problem formulations and property clustering techniques

Decoupling the constitutive equations from the balance and constraint equations allows for reformulating a conventional forward problem into two reverse problems. The first reverse problem is the reverse of a simulation problem, where the process model is solved in terms of the constitutive (synthesis/design) variables instead of the process variables, thus providing the synthesis/design targets. The second reverse problem (reverse property prediction) solves the constitutive equations to identify unit operations, operating conditions and/or products by matching the synthesis/design targets. Visualization of the problem is achieved by employing recently developed property clustering techniques, which allows a high-dimensional problem to be visualized in two or three dimensions. The clusters by definition satisfy intra-stream and inter-stream conservation through linear “mixing” rules, which allows for the development of consistent additive rules along with their ternary representation.     

Parental Problem Drinking and Children's Adjustment: Attachment and Family Functioning as Moderators and Mediators of Risk.

Children's perceived attachments with parents, and family cohesion and adaptability were examined as predictors, mediators, and moderators in the parental problem drinking-child outcomes link. A total of 216 6- to 12-year-olds (110 boys, 106 girls) participated. Data were obtained from children and their mothers, fathers, and teachers. A higher level of family cohesion and adaptability functioned as (a) a robust protective factor against adjustment and cognitive difficulties otherwise associated with problem drinking and (b) a mediator of adjustment problems. Children's perceptions of attachments to mothers and fathers were consistent predictors of behavioral, social, and cognitive problems and further moderated relations between problem drinking and child functioning. The results support that child-parent and family functioning variables act as either pathways and/or vulnerability and protective factors for children exposed to a high-risk environment. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

基于区域生长法的医学图像分割研究

针对基于区域的分水岭分割算法中通常存在的严重过分割现象，通过两点措施来改进：其一，以梯度图像的浮点活动图像取代梯度图像进行分水岭变换，使边缘定位更准确；其二，以基于面积和对比度控制的合并小区域准则，有效抑制过分割现象，同时满足感兴趣区的分割要求。实验表明该算法简单有效，能够得到符合人类视觉系统特性的分割结果，边缘定位精确度较高。     

一个改进的单抓钩周期性排序模型及其在自动化学处理线中的应用

计算机控制的抓钩广泛用于自动化学处理线的工件的运送。抓钩的排序直接影响系统的生产率，抓钩排序的目标是对运送进行排序以极大化生产率。当某工序处理时间非常长时，该工序成为瓶颈。为了去除该瓶颈，系统可以为该工序设计多个处理槽，这称为“多重处理槽”问题。本文提出一个改进的混合整数规划模型以求解有“多重处理槽”的单抓钩周期性排序问题的最优解。实例表明所提出的方法是有效的。     

Hemispheric contributions to nonverbal abstract reasoning and problem solving.

Hemispheric involvement in reasoning abilities has been debated for some time, and it remains unclear whether the right hemisphere's involvement in problem solving is modality specific or dependent on the type of spatial reasoning required. In the current study, 2 types of nonverbal reasoning abilities were examined, spatial reasoning and proportional reasoning, in 109 patients with cerebrovascular disease that was confined to either the right or the left hemisphere or was diffuse in nature. Results indicated that no lateralizing effects were present based on type of spatial reasoning. Findings are consistent with the suggestion that higher order cognitive processes involved in nonverbal abstraction and problem solving are not strongly lateralized to the right hemisphere but rather are more generally distributed throughout the cortex. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Real-world extensions to scheduling algorithms based on lagrangian relaxation

Recently, efficient scheduling algorithms based on Lagrangian relaxation have been proposed for scheduling parallel machine systems and job shops. In this article, we develop real-world extensions to these scheduling methods. In the first part of the paper, we consider the problem of scheduling single operation jobs on parallel identical machines and extend the methodology to handle multiple classes of jobs, taking into account setup times and setup costs. The proposed methodology uses Lagrangian relaxation and simulated annealing in a hybrid framework. In the second part of the paper, we consider a Lagrangian relaxation based method for scheduling job shops and extend it to obtain a scheduling methodology for a real-world flexible manufacturing system with centralized material handling. This research was supported in part by the Office of Naval Research and the Department of Science and Technology grant N00014-93-1017.     

静态SAT问题并行处理器的设计研究

针对SAT问题的复杂性及求解速度缓慢的问题，采用可重构器件FPGA设计，实现了静态回溯搜索算法SAT问题并行处理器，提出了研制动态SAT并行处理器的设想。     

Hierarchical production control for a flow shop with dynamic setup changes and random machine breakdowns

In this paper, we study a manufacturing system consisting of two machines separated by two intermediate buffers, and capable of producing two different products. Each product requires a constant processing time on each of the machines. Each machine requires a constant non-negligible setup change time from one product to the other. The demand rate for each product is considered to be piecewise constant. Each machine undergoes failure and repair. The time-to-failure and time-to-repair are exponentially distributed random variables. The setup change and processing operations are resumable. We model our system as a continuous time, continuous flow process. An optimal control problem is formulated for the system to minimize the total expected discounted cost over an infinite horizon. To determine the optimal control policy structure, a discrete version of the problem is solved numerically using a dynamic programming formulation with a piecewise linear penalty function. A real-time control algorithm is then developed with the objective of maintaining low work-in-process inventory and keeping the production close to the demand. The algorithm uses a hierarchical control structure to generate the loading times for each product on each machine in real time and to respond to random disruptions in the system. The system is simulated using this algorithm to study its performance. The performance of the algorithm is also compared to alternative policies.