A computing framework for integrating interactive visualization in HPCC applications

Network-based concurrent computing and interactive data visualization are two important components in industry applications of high-performance computing and communication. We propose an execution framework to build interactive remote visualization systems for real-world applications on heterogeneous parallel and distributed computers. Using a dataflow model of a commercial visualization software AVS in three case studies, we demonstrate a simple, effective, and modular approach to couple parallel simulation modules into an interactive remote visualization environment. The applications described in this paper are drawn from our industrial projects in financial modeling, computational electromagnetics and computational chemistry. Copyright © 1999 John Wiley & Sons, Ltd.     

Optimal loop storage allocation for argument-fetching dataflow machines

In this paper, we consider the optimal loop scheduling and minimum storage allocation problems based on the argument-fetching dataflow architecture model. Under the argument-fetching model, the result generated by a node is stored in a unique location which is addressable by its successors. The main contribution of this paper includes: for loops containing no loop-carried dependences, we prove that the problem of allocating minimum storage required to support rate-optimal loop scheduling can be solved in polynomial time. The polynomial time algorithm is based on the fact that the constraint matrix in the formulation is totally unimodular. Since the instruction processing unit of an argument-fetching dataflow architecture is very much like a conventional processor architecture without a program counter, the solution of the optimal loop storage allocation problem for the former will also be useful for the latter.     

Modular product design with grouping genetic algorithm—a case study

Modular products are products that fulfill various functions through the combination of distinct modules. These detachable modules are constructed both according to the maximum physical and functional relations among components and maximizing the similarity of specifically modular driving forces. Accordingly, a non-linear programming is proposed to identify separable modules and simultaneously optimize the number of modules. This paper presents a systematic approach to accomplish modular product design in four major phases. Phase 1 is by means of functional and physical interaction analysis to format a component-to-component correlation matrix. Phase 2 is the exploration of design requirements to evaluate the relative importance of each modular driver. In phase 3, non-linear programming is used to formulate the objective function. In the final phase, a heuristic grouping genetic algorithm is adopted to search for the optimal or near-optimal modular architecture. This process and its application are illustrated by a real case of an electrical consumer product provided by an Original Design Manufacturer. The results demonstrate that the designer could direct a new approach to establish product modules according to the relative importance of modular drivers and the interaction among components.     

A new paradigm for real-time parallel storage and recognition of patterns based on a hierarchical organization of associative memories utilizing Walsh function encoding

A new hierarchical Walsh memory which can store and quickly recognize any number of patterns is proposed. A Walsh function based associative memory was found to be capable of storing and recognizing patterns in parallel via purely a software algorithmic technique (namely, without resorting to parallel hardware) while the memory itself only takes a single pattern space of computer memory, due to the Walsh encoding of each pattern. This type of distributed associative memory lends itself to high speed pattern recognition and has been reported earlier in a single memory version. In this paper, the single memory concept has first been extended to a parallel memory module and then to a tree-shaped hierarchy of these parallel modules that are capable of storing and recognizing any number of patterns for practical large scale data applications exemplified by image and speech recognition. The memory hierarchy was built by successively applying k-means clustering to the training data set. In the proposed architecture, the clustered data subsets are stored respectively into a parallel memory module where the module allocation is optimized using the genetic algorithm to realize a minimal implementation of the memory structure. The system can recognize all the training patterns with 100% accuracy and further, can also generalize on similar data. In order to demonstrate its efficacy with large scale real world data, we stored and recognized over 500 faces while at same time, achieving much reduced recognition time and storage space than template matching.     

Evolving modular networks with genetic algorithms: application to nonlinear time series

Abstract: A key problem of modular neural networks is finding the optimal aggregation of the different subtasks (or modules) of the problem at hand. Functional networks provide a partial solution to this problem, since the inter‐module topology is obtained from domain knowledge (functional relationships and symmetries). However, the learning process may be too restrictive in some situations, since the resulting modules (functional units) are assumed to be linear combinations of selected families of functions. In this paper, we present a non‐parametric learning approach for functional networks using feedforward neural networks for approximating the functional modules of the resulting architecture; we also introduce a genetic algorithm for finding the optimal intra‐module topology (the appropriate balance of neurons for the different modules according to the complexity of their respective tasks). Some benchmark examples from nonlinear time‐series prediction are used to illustrate the performance of the algorithm for finding optimal modular network architectures for specific problems.     

Synthesis of a nonrecurrent associative memory model based on anonlinear transformation in the spectral domain

A new nonrecurrent associative memory model is proposed. This model is composed of a nonlinear transformation in the spectral domain followed by the association. The Moore-Penrose pseudoinverse is employed to obtain the least squares optimal solution. Computer simulations are done to evaluate the performance of the model. The simulations use one-dimensional speech signals and two-dimensional head/shoulder images. Comparison of the proposed model with the classical optimal linear associative memory and an optimal nonlinear associative memory is presented.     

Face recognition by applying wavelet subband representation and kernel associative memory

In this paper, we propose an efficient face recognition scheme which has two features: 1) representation of face images by two-dimensional (2D) wavelet subband coefficients and 2) recognition by a modular, personalised classification method based on kernel associative memory models. Compared to PCA projections and low resolution "thumb-nail" image representations, wavelet subband coefficients can efficiently capture substantial facial features while keeping computational complexity low. As there are usually very limited samples, we constructed an associative memory (AM) model for each person and proposed to improve the performance of AM models by kernel methods. Specifically, we first applied kernel transforms to each possible training pair of faces sample and then mapped the high-dimensional feature space back to input space. Our scheme using modular autoassociative memory for face recognition is inspired by the same motivation as using autoencoders for optical character recognition (OCR), for which the advantages has been proven. By associative memory, all the prototypical faces of one particular person are used to reconstruct themselves and the reconstruction error for a probe face image is used to decide if the probe face is from the corresponding person. We carried out extensive experiments on three standard face recognition datasets, the FERET data, the XM2VTS data, and the ORL data. Detailed comparisons with earlier published results are provided and our proposed scheme offers better recognition accuracy on all of the face datasets.     

Multi-level redundancy optimization in series systems

Single-level systems have been considered in redundancy allocation problems. It may be the best policy in some specific situations, but not in general. In regards to reliability, it is most effective to duplicate the lowest objects, because parallel-series systems are more reliable than series-parallel systems. However, the smaller an object is, the more time and higher accuracy are needed for duplicating it, and so, redundancy cost can be decreased by using modular redundancy. Therefore, providing redundancy at high levels like as modules or subsystems, can be more economical than providing redundancy at low level of components. In this paper, the problem in which redundancy is available at all levels in a series system is addressed and a mixed integer programming model is presented. A heuristic algorithm and a genetic algorithm are proposed to solve the problem and some examples illustrate the procedure.     

Carrying on the legacy of imperative languages in the future parallel computing era

There has been a renewed interest in dataflow computing models in recent years of technology scaling. Potentiality of exploiting huge parallelism, with the expense of low power, simpler circuit, less silicon area, is the main characteristic of a dataflow model. Growing trends in housing large number of functional units in a single chip, making use of local clocks, reducing energy consumptions, avoiding global wires are the main reasons behind the resurgence of dataflow models. To program a dataflow machine, new architectures suggest imperative languages rather than functional type dataflow languages or parallel languages because this is the right way to make the new architectures popular among the general community. Although for several decades scientists have been working on how imperative languages can be used in dataflow models efficiently, there is no systematic review on those works. Existing reviews on dataflow paradigm mainly focus on the architectures. Although few papers review programming languages of dataflow architectures, their discussions are limited to only dataflow languages and visual programming languages which are fundamentally different from imperative languages. In this paper, we conduct a systematic review on those works that attempt to provide a way to use imperative languages in any type of dataflow architectures. Our survey of compilers and related architectures cover the aspects like translation mechanisms of program construct, their optimization techniques, memory ordering methods, program allocation and scheduling and special architectural features. We also present some of our observations and future research directions obtained by exploring the literature.     

Supervisory Control Architecture for Discrete-Event Systems

A flexible decentralized and hierarchical architecture is presented to reduce computational effort in designing optimal nonblocking supervisors for discrete-event systems (DES). We organize a DES into modular subsystems that embody internal interacting dependencies. Verification of, and coordination among modular subsystems are achieved through their model abstractions. Sufficient conditions are presented to guarantee that coordinators and modular supervisors result in maximally permissive and nonblocking control. A medium-sized example demonstrates the computational effectiveness of our approach.      

具有期望容错域的前向掩蔽联想记忆模型的设计方法

联想记忆的综合问题是目前没有很好解决的难题.文中用作者提出的通用前馈网络和排序学习算法,提出了一种设计具有期望容错域的前向掩蔽联想记忆模型的方法.该方法一般性地解决了信息空间上联想记忆的综合难题,使设计出的联想记忆模型具有任意期望的记忆样本容错域.     

F1owS:一种MapReduce数据流公平调度方法

MapReduce Job的调度机制一直是学术研究的热点。在分析MapReduce数据流调度模型的基础上,提出一种面向MapReduce数据流的公平调度方法FlowS。该方法采用数据流池来分配资源以保证MapReduce数据流的隔离性,并且采用数据流池动态构建算法来确保资源的公平分配。实验表明,该调度方法可以有效提高Hadoop集群对MapReduce数据流的处理效率。     

Parallel modular exponentiation using load balancing without precomputation

The modular exponentiation operation of the current algorithms for asymmetric cryptography is the most expensive part in terms of computational cost. The RSA algorithm, for example, uses the modular exponentiation algorithm in encryption and decryption procedure. Thus, the overall performance of those asymmetric cryptosystems depends heavily on the performance of the specific algorithm used for modular exponentiation. This work proposes new parallel algorithms to perform this arithmetical operation and determines the optimal number of processors that yields the greatest speedup. The optimal number is obtained by balancing the processing load evenly among the processors. Practical implementations are also performed to evaluate the theoretical proposals.     

任务分配与调度的共同进化方法

并行与分布式计算环境中随着独立任务的增多,传统进化类单种群的任务分配与调度算法的效率与效力随之大为降低,该文在分析传统解完整编码单种群进化类算法的基础上,基于生物界多物种间共同进化的机制提出了任务分配与调度的合作式共同进化计算模型,并探讨了任务分配与调度问题中的子种群合作方式与个体的适应值计算方法。此外,从数学上分析了基于合作式共同进化的任务分配与调度算法的性能,指出共同进化调度方法中好的调度方案能以高于传统单种群进化算法的递增指数递增。仿真分析证实了算法的理论分析结果,算法具有实际工程价值。     

Design for Cost: Module-Based Mass Customization

The assemble-to-order (ATO) production strategy considers a tradeoff between the size of a product portfolio and the assembly lead time. The concept of modular design is often used in support of the ATO strategy. Modular design impacts the assembly of products and the supply chain, in particular, the storage, transport, and production are affected by the selected modular structure. The demand for products in a product family impacts the cost of the supply chain. Based on the demand patterns, a mix of modules and their stock are determined by solving an integer programming model. This model cannot be optimally solved due to its high computational complexity and, therefore, two heuristic algorithms are proposed. A simulated annealing algorithm improves on the previously generated solutions. The computational results reported in this paper show that significant savings could be realized by optimizing the composition of modules. The best performance is obtained by a simulated annealing combined with a heuristic approach.     

考虑同质设备不同加工能力的模块化制造单元资源配置优化

针对传统制造单元易出现在制品的积压、加工效率低下的问题,提出一种模块化制造单元资源配置模型及算法。按照模块化思想进行制造单元划分,为了保证单元内部均衡,工作站内可配置若干个同类设备。考虑到同类设备之间的加工效率不同,建立以总加工成本最小及制造单元内与单元间的均衡为目标的数学模型,应用模拟植物生长算法进行求解。最后通过算例验证了模型与算法的有效性。     

一种基于服务选取的SBS云资源优化分配方法

获取满足全局优化目标的资源分配策略,是影响云环境中基于服务的软件系统(service-based software system,简称SBS)运行时优化效果的关键.然而,由于SBS内部复杂的业务逻辑关系和云环境中的资源约束,现有分配方法无法得到最优资源分配量.以满足SLA约束和最小化资源成本为目标,根据不同资源状态对应不同组件服务性能的特点,将组件服务可能的资源分配量、相应性能及成本转换为备选逻辑服务集,进而提出了一种云环境中基于服务选取的SBS资源优化分配模型,并设计了一种求解模型的混合遗传算法.算法采用整数编码以提高求解效率,并在选择算子中引入了精英保留策略,从而保证收敛到全局最优解.为提高遗传算法的局部搜索能力、加快收敛速度,以局部搜索策略改进了标准变异算子.实验验证了所提出的资源优化分配模型和求解算法的有效性,并表明:与分支定界法及精英保留策略遗传算法相比,混合遗传算法能够在较大规模的问题上快速获得具有较低资源成本的资源分配策略.     

SPASS: a generic process-oriented crop model with versatile windows interfaces

Crop simulation models are now widely used in agronomy. A highly modular structure has been developed for crop growth simulation models based on the essential processes that are common to various plant species. This generic process-oriented modular structure was implemented in the SPASS (Soil-Plant-Atmosphere System Simulation) model using Microsoft Visual C/C++ under Windows. SPASS is composed of six parts: (i) data management, (ii) memory allocation/initialization, (iii) Modules for crop, soil and microclimate processes, (iv) model configuration and simulation, (v) graphic/table display and (vi) model knowledge integration. The model can be easily configured using the Windows interface by choosing various combinations of process modules developed based on different model approaches. Up to 30 graphic and 5 table MDI (Multiple Document Interface) child windows can be opened during or after the simulation run. Each child window can be resized freely on the screen for the convenience of data analysis. The model theory can be easily viewed using the online theory book. The SPASS model structure facilitates comparison of different model approaches, transfer of improvements between models, development of multi-crop models and modification of process modules for new applications.