A workflow system is a set of computer-supported activities organized as a single system under various flow patterns. The time management of a workflow system concerns the modelling of schedules of individual activiti...
详细信息
ISBN:
(纸本)088986392X
A workflow system is a set of computer-supported activities organized as a single system under various flow patterns. The time management of a workflow system concerns the modelling of schedules of individual activities and the testing of compatibility among the schedules of different activities in centralized or distributed environment. In this work, based on Windows-Explorer-like workflow process model, we propose a more flexible schedule format named schedule frame, and develop a new concept of time-compatibility which compromises the two extreme aspects of prevailing ones and is defined in terms of feasible conditions. In the execution level, the algorithm for maintaining time-compatibility is presented.
The interacting multiple model (IMM) algorithm is parallelised using combined task and data parallelism technique and is represented by an acyclic task graph. The tasks of the parallel IMM algorithm, are scheduled ont...
详细信息
ISBN:
(纸本)088986392X
The interacting multiple model (IMM) algorithm is parallelised using combined task and data parallelism technique and is represented by an acyclic task graph. The tasks of the parallel IMM algorithm, are scheduled onto a multiprocessor system using the Data Token Heuristic (DTH) approach. The completion time of the parallel IMM is studied on a simulated transputer network, depicting the message-passing multicomputer system, for a varying number of processors on different interconnection topologies.
In this paper, we introduce an easy-to-use, platform senseless task parallel programming model called Lily Task, as well as its implementations on SMP and Cluster. Lily Task programming model reflects directly the hum...
详细信息
ISBN:
(纸本)088986392X
In this paper, we introduce an easy-to-use, platform senseless task parallel programming model called Lily Task, as well as its implementations on SMP and Cluster. Lily Task programming model reflects directly the human thoughts in the parallelism process of a problem. LilyTask supports both data and function parallelisms, as well as dynamic task creation and scheduling. It is very suitable for solving irregular problems. And its performance is satisfying.
On parallel computer systems which need communications and synchronizations, granularity of parallel execution and data locality are important for performance of a program. The most well-known program transformation w...
详细信息
ISBN:
(纸本)088986392X
On parallel computer systems which need communications and synchronizations, granularity of parallel execution and data locality are important for performance of a program. The most well-known program transformation which improves data locality and finds coarse-grain parallelism is tiling. However, it is a sequential program code that is generated by conventional methods and then the code is further transformed for parallel processing. We propose a tiled code generation method which combines the two phases of conventional methods, tiled code generation and parallelization. We show that the proposed method is convenient to the optimization of DOALL parallel processing for minimizing synchronization.
Grid coteries are attractive both in message and space complexities, but they are not non-dominated (ND) unfortunately. This paper introduces the transversal merge operation for constructing an ND coterie from a domin...
详细信息
ISBN:
(纸本)088986392X
Grid coteries are attractive both in message and space complexities, but they are not non-dominated (ND) unfortunately. This paper introduces the transversal merge operation for constructing an ND coterie from a dominated one, and applies it to grid coteries to make them ND. The constructed ND grid coteries sill preserve favorite features that the original grid coteries have. To demonstrate this fact, we evaluate their quorum sizes and design a mutual exclusion algorithm based on the ND grid coteries, which allows a process to construct a quorum on the fly.
List-based scheduling is generally accepted as an attractive approach to static task scheduling in homogeneous environment, since it pairs low complexity with good results. This paper presents a static list-scheduling...
详细信息
ISBN:
(纸本)088986392X
List-based scheduling is generally accepted as an attractive approach to static task scheduling in homogeneous environment, since it pairs low complexity with good results. This paper presents a static list-scheduling algorithm: Critical Path Parent Trees (CPPT) algorithm. The CPPT algorithm divides the task graph into a set of unlisted parenttrees, the root of each parent-tree is a critical-path node. The analysis and experiments have shown that the algorithm provides very low complexity and comparable results.
In this paper, we show how the use of ontologies has helped us address some key issues facing ubiquitous computing environments;improving interoperability between entities, discovery and matchmaking, and context aware...
详细信息
ISBN:
(纸本)088986392X
In this paper, we show how the use of ontologies has helped us address some key issues facing ubiquitous computing environments;improving interoperability between entities, discovery and matchmaking, and context awareness. Ontologies establish a joint terminology between members of a community of interest and provide a basis for achieving semantic interoperability. We have integrated the use of ontologies in our Smart Spaces framework GAIA [10, 13]. We show that ontologies provide important capabilities for Ubiquitous computing infrastructure, and context-aware computing.
An important problem in the construction of distributedsystems is the openness of these systems to dynamic computations, i.e. the ability of these systems to support new infrastructure services and applications durin...
详细信息
ISBN:
(纸本)088986392X
An important problem in the construction of distributedsystems is the openness of these systems to dynamic computations, i.e. the ability of these systems to support new infrastructure services and applications during their operations. Algorithms and services have been designed using several system models: synchronous, asynchronous bonded-delay, asynchronous, etc. In practice, most of the existing systems implement a limited number of system models, and can be hardly adapted to new applications. For example, a metropolitan area network used to develop a reliable data communication system can hardly support real-time audio-communications. In this paper, we describe new principles and an architectural basis for the development of distributed system infrastructures supporting multiple system models, that can be dynamically modified to be adapted to new application requirements, using programmable networks. The paper provides a framework for the application of network programmability and active networks to distributed system design, and their use to dynamically enforce multiple and concurrent system models. Such systems can be used to support multiple application profiles with the same computing and communication resources.
In a series of recent papers, we showed how to avoid the impossibility result of deterministic consensus in presence of link failures [Gray,1978] for well-known Oral and Written Messages Byzantine agreement algorithms...
详细信息
ISBN:
(纸本)088986392X
In a series of recent papers, we showed how to avoid the impossibility result of deterministic consensus in presence of link failures [Gray,1978] for well-known Oral and Written Messages Byzantine agreement algorithms. This is done by employing the perception-based fault model [Schmid,2001]. We will, however, use a simplified version of that model in this paper. The Byzantine Agreement protocol presented in this paper is a polynomial time protocol with optimal resilience in respect to the simplified perception-based fault model. The most important feature of the Byzantine Agreement algorithm is, that it uses some of the processes to tolerate process failures as well as to overcome link failures, as we need only n > 2fa + max(fa, f lr,a + flr + fl s,a + fls), where flr, flr,a, fls and fl s,a are measures of the link failures allowed.
We describe a technique for executing highly-parallel programs on a network and a prototype system that uses the technique. The programs are written in the Lucid family of programming languages. Lucid is a multidimens...
详细信息
ISBN:
(纸本)088986392X
We describe a technique for executing highly-parallel programs on a network and a prototype system that uses the technique. The programs are written in the Lucid family of programming languages. Lucid is a multidimensional declarative programming language in which computations are performed in an implicit evaluation context. The execution engine is using a demand-driven execution model and operates in a client/server environment. High-level Lucid code is translated at compile-time into an object-oriented representation in which each Lucid identifier is represented by a class and each Lucid value (identifier in a context) is an instance of this class. The execution engine uses a cache to improve the overall performance and is designed to achieve both flexibility and efficiency.
暂无评论