On the history of Ludwig von Bertalanffy’s “general systemology”, and on its relationship to cybernetics - Part II: Contexts and developments of the systemological hermeneutics instigated by von Bertalanffy

The history of “general system theory” is investigated in order to clarify its meanings, vocations, foundations and achievements. It is characterized as the project of a science of the systemic interpretation of the “real”, renamed here “general systemology”. The contexts and modes of its elaboration, publication and implementation are discussed. The paper mostly focuses on the works of its instigator: Ludwig von Bertalanffy. However, general systemology was a collective project: the main contributions of other “systemologists”, from the 1950s until the 1970s, are hence also considered. Its solidarity with the history of the Society for General Systems Research is notably discussed. A reconstruction of the systemological hermeneutics is undertaken on this basis. It finds out the potential systematic unity underlying the diversity of the contributions to this both scientific and philosophical project. Light is thus shed on the actual scope of von Bertalanffy’s works.     

Kokkos: Enabling manycore performance portability through polymorphic memory access patterns

The manycore revolution can be characterized by increasing thread counts, decreasing memory per thread, and diversity of continually evolving manycore architectures. High performance computing (HPC) applications and libraries must exploit increasingly finer levels of parallelism within their codes to sustain scalability on these devices. A major obstacle to performance portability is the diverse and conflicting set of constraints on memory access patterns across devices. Contemporary portable programming models address manycore parallelism (e.g., OpenMP, OpenACC, OpenCL) but fail to address memory access patterns. The Kokkos C++ library enables applications and domain libraries to achieve performance portability on diverse manycore architectures by unifying abstractions for both fine-grain data parallelism and memory access patterns. In this paper we describe Kokkos’ abstractions, summarize its application programmer interface (API), present performance results for unit-test kernels and mini-applications, and outline an incremental strategy for migrating legacy C++ codes to Kokkos. The Kokkos library is under active research and development to incorporate capabilities from new generations of manycore architectures, and to address a growing list of applications and domain libraries.     

BiELL: A bisection ELLPACK-based storage format for optimizing SpMV on GPUs

Sparse matrix–vector multiplication (SpMV) is one of the most important high level operations for basic linear algebra. Nowadays, the GPU has evolved into a highly parallel coprocessor which is suited to compute-intensive, highly parallel computation. Achieving high performance of SpMV on GPUs is relatively challenging, especially when the matrix has no specific structure. For these general sparse matrices, a new data structure based on the bisection ELLPACK format, BiELL, is designed to realize the load balance better, and thus improve the performance of the SpMV. Besides, based on the same idea of JAD format, the BiJAD format can be obtained. Experimental results on various matrices show that the BiELL and BiJAD formats perform better than other similar formats, especially when the number of non-zero elements per row varies a lot.     

图的点可区别边染色算法研究

针对一般图设计了一种新型的点可区别边染色算法。该算法把概率思想和图染色相结合, 根据点可区别边染色的约束规则确立目标函数, 利用交换规则逐步寻优, 当目标函数的值满足要求时染色成功。给出详细算法步骤并进行了测试和分析, 实验结果表明该算法可以求出满足猜想的点可区别边色数。     

面向大规模数据流处理的偏最小二乘法

用光谱分析鉴别生物特征,导致数据量大,而实际需要必须实时处理。偏最小二乘法是使用最广泛的鉴别算法,但是对于大规模数据流该算法无法达到实时性。为了解决这个应用矛盾,提出了一种基于NVIDIA CUDA架构下的并行计算策略,利用具有大规模并行计算特征的图形处理器（GPU）作为计算设备,结合GPU存储器的优势实现了偏最小二乘算法。实验的测试结果表明,在GPU上使用CUDA实现的偏最小二乘算法比在CPU上实现该算法快了47倍,性能得到了显著提高,从而使偏最小二乘算法应用于大规模数据流处理成为可能。     

基于PXI技术的飞控计算机通用测试平台的构建

针对飞控计算机的信号特性及数量,构建以PXI总线技术为核心的通用测试设备,能够满足多余度计算机的测试。对于不同型号的飞控计算机只需开发不同的测试适配器和信号调理板,就能满足不同型号的测试要求,达到资源共享、快速研制的目的。该通用平台同时可以满足产品从研制、试验、到小批量生产的要求。     

基于网络统考课程的学习支持服务构建

为满足学生在网络统考课程学习中自主学习的需要,学习支持服务系统的研究、设计构建和有效运行至关重要。各试点高校关于网络统考学习支持服务的实践和做法很多,但系统性研究和理论总结不够,作者结合工作实际尝试构建了网络统考学习支持服务体系基本框架。     

图形处理器片段处理单元的设计与实现

针对图形处理器三维引擎中对图形的后期处理需求,实现片段写入帧缓冲区前的测试、混合、逻辑操作、累积、清除和屏蔽等关键功能。分析并提取了OpenGL核心库中的片段处理相关函数,确定了片段处理单元要实现的功能;合理安排多个片段处理功能的执行顺序,设计了基于流水线的片段处理单元结构;采用Verilog HDL对电路进行描述,采用 Cadence NC-Verilog 仿真工具进行虚拟验证,采用 Xilinx 的 ISE 工具进行综合,并在 Xilinx Virtex6 XC6VLX760 FPGA上进行原型验证,电路工作频率可以达到180 MHz,测试功能正确。在SMIC 65 nm CMOS工艺下,采用Synopsys Design-Compiler对设计进行综合,电路工作频率达到300 MHz,满足设计需求。     

基于CUDA架构的FFT并行计算研究

FFT（快速傅里叶变换）是基于提高DFT（离散傅里叶变换）计算的高效算法,它在众多科学和工程领域都得到了广泛的应用。自FFT算法出现以后,从早期的以降低复杂度到近年以来的大规模并行FFT计算,各种优化算法得到广泛的研究。在并行运算领域中,随着可编程的、并行化GPU的不断推广,特别是通用并行统一计算架构CUDA的出现,极大增强了GPU的计算能力,在编程和优化等方面都有显著地提升。鉴于此,本文在分析FFT算法实现的基础上,研究了一种适合GPU运算的FFT并行计算方法,并通过CUDA架构实现了FFT算法在GPU上的运算。该方法的引入在理论不计算数据传输的情况下,使一维FFT运算时间的复杂度由O（N logN2）可以降到O（N/rlogN2）。通过验证,本文提出的CUDA的并行FFT方法得到较好的加速效果,在精度计算上也符合实际的要求,从而证明了该方法的正确性和有效性。     

Conceptual design of a novel hybrid fuel cell/desalination system

A novel concept for integrating fuel cells with desalination systems is proposed and investigated in this work. Two unique case studies are discussed — the first involving a hybrid system with a reverse osmosis (RO) unit and the second — integrating with a thermal desalination process such as multi-stage flash (MSF). The underlying motivation for this system integration is that the exhaust gas from a hybrid power plant (fuel cell/turbine system) contains considerable amount of thermal energy, which may be utilized for desalination units. This exhaust heat can be suitably used for preheating the feed in desalination processes such as reverse osmosis which not only increases the potable water production, but also decreases the relative energy consumption by approximately 8% when there is an increase of just 8°C rise in temperature. Additionally, an attractive hybrid system application which combines power generation at 70%+ system efficiency with efficient waste heat utilization is thermal desalination. In this work, it is shown that the system efficiency can be raised appreciably when a high-temperature fuel cell co-generates DC power in-situ with waste heat suitable for MSF. Results indicate that such hybrid system could show a 5.6% increase in global efficiency. Such combined hybrid systems have overall system efficiencies (second-law base) exceeding those of either fuel-cell power plants or traditional desalination plants.