Adapting system execution traces to support analysis of software system performance properties

UNITE is a method and tool that analyzes software system performance properties, e.g., end-to-end response time, throughput, and service time, via system execution traces. UNITE, however, assumes that a system execution trace contains properties (e.g., identifiable keywords, unique message instances, and enough variation among the same event types) to support performance analysis. With proper planning, it is possible to ensure that properties required to support such analysis are incorporated in the generated system execution trace. It, however, is not safe to assume this to be the case with many existing software systems.     

并行原型系统上BFS算法设计实现与测试分析

相对于传统应用,大数据应用表现出并行性高、访存数据量大、访存模式不规则、程序访存时空局部性差等特性,对传统的计算机体系结构提出了新的挑战。Graph500是评测计算机系统大数据处理能力的基准测试排名,BFS算法是Graph500的核心程序,是典型的数据密集型应用。从1-D数据划分、优化的混合算法设计和远程通信方式设计三个方面开展研究,在课题组设计的大数据处理并行结构原型系统上设计实现了多节点的并行BFS算法,在222顶点、226边的数据规模下取得了803.8MTEPS的性能,并在此基础上进行多节点并行BFS算法的性能测试分析,为进一步的研究工作奠定了基础。     

User transparency: a fully sequential programming model for efficient data parallel image processing

Although many image processing applications are ideally suited for parallel implementation, most researchers in imaging do not benefit from high‐performance computing on a daily basis. Essentially, this is due to the fact that no parallelization tools exist that truly match the image processing researcher's frame of reference. As it is unrealistic to expect imaging researchers to become experts in parallel computing, tools must be provided to allow them to develop high‐performance applications in a highly familiar manner. In an attempt to provide such a tool, we have designed a software architecture that allows transparent (i.e. sequential) implementation of data parallel imaging applications for execution on homogeneous distributed memory MIMD‐style multicomputers. This paper presents an extensive overview of the design rationale behind the software architecture, and gives an assessment of the architecture's effectiveness in providing significant performance gains. In particular, we describe the implementation and automatic parallelization of three well‐known example applications that contain many fundamental imaging operations: (1) template matching; (2) multi‐baseline stereo vision; and (3) line detection. Based on experimental results we conclude that our software architecture constitutes a powerful and user‐friendly tool for obtaining high performance in many important image processing research areas. Copyright © 2004 John Wiley & Sons, Ltd.     

流式处理系统的动态数据分配技术

流式数据处理中,数据倾斜等原因易导致计算节点的负载不均衡,降低系统处理能力。传统的负载均衡方法,比如算子分配、算子迁移和负载脱落等技术因为相对较高的性能代价,在流式处理系统中没有得到广泛的应用。针对流式处理系统的特点,提出一种新的负载均衡方法。在该方法中,计算单元的数据被划分为若干分区,并且数据分区可以在计算单元中动态分配和迁移,在较少干扰系统运行的情况下,通过动态调整各计算单元的分区,平衡各个计算单元的输入流和利用率,以此达到负载平衡的目的。在此基础上,设计并实现了流式处理系统的负载均衡算法和数据在线迁移技术。实验结果表明,该方法能够显著减少数据处理的平均延迟,提高系统吞吐量。     

GPROC—an integrated system for the processing of numerical scientific data

Work done on an interactive graphical processing system is described. An overview is given of the uses to which a system could be put, with especial reference to aspects of solid state physics. Algorithms which may be usefully included in such a system, when used in a technical scientific environment, are described. The problems of implementing such a system on small microcomputer systems are outlined, and some possible areas of future development are discussed.     

行业统计报表处理系统SRAFT的设计与实现

分析了行业统计工作中报表制度编制和报表数据处理的特点，给出了一个计算及审核公式的描述方法，设计并实现了一个由报表框架设计器、计算及审核公式管理器、数据汇总项目管理器、填报单位管理器、报表数据处理器等5个部分组成的统计报表处理系统SRAFT。     

Observations of a firm's information processing with a data base management system

The organization in this study used all versions of IMS, from IMS 1 through IMS VS 1.01, over the period 1970–1975. During this period, the number of messages processed steadily increased from 6000 in October 1970 to a maximum in excess of 150,000 in September 1974. This volume enhancement was achieved by solving a sequence of problems concerning IMS software, data base design, or program coding. Perhaps the most important factors supporting the increase were two improvements in IMS software: one was the data base buffer pool introduced with IMS 2; the other the feature introduced with IMS VS that, unlike the previous versions, enabled the simultaneous updating of different data base segments belonging to the same type.     

方差分析数据处理系统的设计

利用Visual Basic设计了方差分析数据处理系统。简单介绍了方差分析的概念及作用,详细介绍了方差分析数据处理系统的设计,并对所设计的数据处理系统进行了性能评价。用方差分析数据处理系统考察影响油页岩渣浸出过程,发现B(碱液浓度)(F=73.1616)和C(油页岩渣粒径)(F=82.0762)是影响浸出过程的主要影响因素。表明本系统能在化学工业生产领域发挥良好的作用。     

Operator-aware approach for boosting performance in RDF stream processing

To enable efficiency in stream processing, the evaluation of a query is usually performed over bounded parts of (potentially) unbounded streams, i.e., processing windows “slide” over the streams. To avoid inefficient re-evaluations of already evaluated parts of a stream in respect to a query, incremental evaluation strategies are applied, i.e., the query results are obtained incrementally from the result set of the preceding processing state without having to re-evaluate all input buffers. This method is highly efficient but it comes at the cost of having to maintain processing state, which is not trivial, and may defeat performance advantages of the incremental evaluation strategy. In the context of RDF streams the problem is further aggravated by the hard-to-predict evolution of the structure of RDF graphs over time and the application of sub-optimal implementation approaches, e.g., using relational technologies for storing data and processing states which incur significant performance drawbacks for graph-based query patterns. To address these performance problems, this paper proposes a set of novel operator-aware data structures coupled with incremental evaluation algorithms which outperform the counterparts of relational stream processing systems. This claim is demonstrated through extensive experimental results on both simulated and real datasets.