JAPS is an automatic parallelizing system based on JAVA running on NOW. It implements the automatic process from dependence analysis to parallel execution. The current version of JAPS can exploit functional parallelis...
详细信息
JAPS is an automatic parallelizing system based on JAVA running on NOW. It implements the automatic process from dependence analysis to parallel execution. The current version of JAPS can exploit functional parallelism and the detection of data parallelism will be incorporated in the new version, which is underway. The framework and key techniques of JAPS are presented. Specific topics discussed are task partitioning, summary information collection, data dependence analysis, pre-scheduling and dynamic scheduling, etc.
This paper introduces a new model for inter-layer message communication called object mapping model that achieves separation between message preparation and destination selection. In this model, as opposed to traditio...
详细信息
presents a comprehensive and systematic discussion of theoretics, methods, techniques and tools for performance evaluation, prediction and visualization of parallel systems. Chapter 1 gives a short overview of perf...
详细信息
ISBN:
(数字)9781461551478
ISBN:
(纸本)9780792384625;9781461373438
presents a comprehensive and systematic discussion of theoretics, methods, techniques and tools for performance evaluation, prediction and visualization of parallel systems. Chapter 1 gives a short overview of performance degradation of parallel systems, and presents a general discussion on the importance of performance evaluation, prediction and visualization of parallel systems. Chapter 2 analyzes and defines several kinds of serial and parallel runtime, points out some of the weaknesses of parallel speedup metrics, and discusses how to improve and generalize them. Chapter 3 describes formal definitions of scalability, addresses the basic metrics affecting the scalability of parallel systems, discusses scalability of parallel systems from three aspects: parallel architecture, parallel algorithm and parallel algorithm-architecture combinations, and analyzes the relations of scalability and speedup. Chapter 4 discusses the methodology of performance measurement, describes the benchmark- oriented performance test and analysis and how to measure speedup and scalability in practice. Chapter 5 analyzes the difficulties in performance prediction, discusses application-oriented and architecture-oriented performance prediction and how to predict speedup and scalability in practice. Chapter 6 discusses performance visualization techniques and tools for parallel systems from three stages: performance data collection, performance data filtering and performance data visualization, and classifies the existing performance visualization tools. Chapter 7 describes parallel compiling-based, search-based and knowledge-based performance debugging, which assists programmers to optimize the strategy or algorithm in their parallel programs, and presents visual programming-based performance debugging to help programmers identify the location and cause of the performanceproblem. It also provides concrete suggestions on how to modify their parallel program to
Agents being a kind of restricted intelligent objects, it is a natuml way to research the inheritance feature of software agents and integmte inheritance mechanisms into AOP. Based on BDI model of agents, this paper a...
详细信息
The paper introduces a new model for inter-layer message communication called object mapping model that achieves separation between message preparation and destination selection. In this model, as opposed to tradition...
详细信息
The paper introduces a new model for inter-layer message communication called object mapping model that achieves separation between message preparation and destination selection. In this model, as opposed to traditional message delivery model, sender objects of messages need not indicate the destination objects of the messages. The messages can be intercepted by special objects called mapper objects. When mapper objects intercept messages, they can decide to which objects the messages should be sent. Using this model, layered cooperative systems can be realized by using objects to construct each layer and using mapper objects to realize object mapping relationship, and object mapping model has many advantages such as modularity, transparency, implicit invocation and dynamic mapping.
With agents being a kind of restricted intelligent object, it is a natural way to research the inheritance feature of software agents and integrate inheritance mechanisms into AOP. Based on BDI models of agents, the p...
详细信息
With agents being a kind of restricted intelligent object, it is a natural way to research the inheritance feature of software agents and integrate inheritance mechanisms into AOP. Based on BDI models of agents, the paper addresses the semantics of static inheritance and dynamic inheritance of agents. For static inheritance, the semantics are discussed from two perspectives: single inheritance and multiple inheritance. For dynamic inheritance, we identify and formally classify dynamic cloning behaviors of agent instances into four types: function split, logic split, preference split and retrogress split.
The Neocognitron, inspired by the mammalian visual system, is a complex neural network with numerous parameters and weights which should be trained in order to utilise it for pattern recognition. However, it is not ea...
详细信息
For real-time systems, UML sequence diagrams describe interaction among objects, which show the scenarios of system behaviour. In this paper, we give the solution for timing analysis of simple UML sequence diagrams wh...
ISBN:
(纸本)9783540667124
For real-time systems, UML sequence diagrams describe interaction among objects, which show the scenarios of system behaviour. In this paper, we give the solution for timing analysis of simple UML sequence diagrams which describe exactly one scenario without any alternatives and loops, and develop an algorithm for checking the compositions of UML sequence diagrams, which describe multiple scenarios, for timing consistency.
This paper proposes a knowledge sharing and collaboration system model (InKB) based on the Internet. It gives a way to collaborate and share knowledge between Web-based knowledge systems. InKB model has three layers: ...
详细信息
This paper proposes a knowledge sharing and collaboration system model (InKB) based on the Internet. It gives a way to collaborate and share knowledge between Web-based knowledge systems. InKB model has three layers: data exchange layer, collaboration layer and knowledge-based application layer. The data exchange layer solves the problem of how to represent and manipulate knowledge. The collaboration layer devotes collaboration between servers over the Internet. The knowledge-based application layer defines the user interface for knowledge processing over the Internet, such as information searching, decision-support application and data mining. Data exchange between InKB Webs is in XML format, knowledge can be shared between heterogeneous knowledge bases, and knowledge can be remotely manipulated. Collaboration between InKBs is supported through a collaboration agent, which can find the resource the user wants in the InKB system. As an open system model, InKB can also support HTTP requests such as information retrieval and browsing.
暂无评论