ER nets are an extension of Petri nets and suitable for finding solutions to a wide spectrum of problems. The author's idea is to use ER nets not only in the early stages of software development (e.g. specificatio...
详细信息
ER nets are an extension of Petri nets and suitable for finding solutions to a wide spectrum of problems. The author's idea is to use ER nets not only in the early stages of software development (e.g. specification and simulation), but also for real application-execution. The author suggests the use of a virtual computer, based upon ER formalism, ensuring an adequate environment for the entire application life cycle. The reason for such a decision was the good results obtained in Tools Manager development (J. Strucl, 1990), within the framework of the NETwork project (M. Braga et al., 1990). The high-level Petri nets represent parallel tasks in a very intuitive way and thus are regarded as natural tools for solving a wide range of problems. The vision is an environment for the work with ER nets where these are used for problem specification, simulation, prototype building, for final program, support tools, the environment in which the individual stages are not separated, but they are intermingled and represent an entity. The prototype may thus at the same time be the final program, and specification also the documentation.< >
This paper advocates a configuration approach to parallel programming for distributed memory multicomputers, in particular, arrays of transputers. The configuration approach prescribes the rigorous separation of the l...
详细信息
This paper advocates a configuration approach to parallel programming for distributed memory multicomputers, in particular, arrays of transputers. The configuration approach prescribes the rigorous separation of the logical structure of a program from its component parts. In the context of parallel programs, components are processes which communicate by exchanging messages. The configuration defines the instances of these processes which exist in the program and the paths by which they are interconnected. The approach is demonstrated by a toolset (Tonic) which embodies the configuration paradigm. A separate configuration language is used to describe both the logical structure of the parallel program and the physical structure of the target multicomputer. Different logical to physical mappings can be obtained by applying different physical configurations to the same logical configuration. The toolset has been developed from the Conic system for distributed programming. The use of the toolset is illustrated through its application to the development of a parallel program to compute Mandelbrot sets.
The notions of `loosely-coupled' and `tightly-coupled' processes have been used in an intuitive way for many years by system designers. In this paper, we propose a precise characterization. All point-to-point ...
详细信息
The notions of `loosely-coupled' and `tightly-coupled' processes have been used in an intuitive way for many years by system designers. In this paper, we propose a precise characterization. All point-to-point message-communicating processes are loosely-coupled under our characterization. Furthermore, a number of common paradigms of process interactions - e.g. broadcast or accumulating a result - satisfy our characterization of loose-coupling. Restricting our designs to loosely-coupled processes has several advantages, among them a design methodology in which each process implements a progress property and a simple solution to the cache coherence problem.
The correctness of most randomized distributed algorithms is expressed by a statement of the form `some predicate of the executions holds with high probability, regardless of the order in which actions are scheduled&#...
详细信息
ISBN:
(纸本)0897914953
The correctness of most randomized distributed algorithms is expressed by a statement of the form `some predicate of the executions holds with high probability, regardless of the order in which actions are scheduled'. In this paper, we present a general methodology to prove correctness statements of such randomized algorithms. Specifically, we show how to prove such statements by a series of refinements, which terminate in a statement independent of the schedule. To demonstrate the subtlety of the issues involved in this type of analysis, we focus on Rabin's randomized distributed algorithm for mutual exclusion. Surprisingly, it turns out that the algorithm does not maintain one of the requirements of the problem under a certain schedule. In particular, we give a schedule under which a set of processes can suffer lockout for arbitrary long periods.
The problem of fault-tolerant coordination is fundamental in distributed computing. In the past, researchers have considered two types of coordination: general coordination, in which the actions of faulty processors a...
详细信息
ISBN:
(纸本)0897914953
The problem of fault-tolerant coordination is fundamental in distributed computing. In the past, researchers have considered two types of coordination: general coordination, in which the actions of faulty processors are irrelevant, and consistent coordination, in which the faulty processors are forbidden from acting inconsistently. This paper studies the possibility and complexity of achieving coordination in synchronous and asynchronous systems with crash, send-omission, and general omission failures. We indicate the systems in which coordination cannot be achieved and, when it can, analyze the computational complexity of optimally achieving it. In some cases, optimum solutions can be implemented in polynomial time, while in others they require NP-hard local computation. These results provide a thorough characterization of coordination and will thus aid researchers in determining the approach to take when attempting to achieve fault-tolerant coordination.
Heterogeneity has been recognized as an important way of improving the performance of multiprocessors by using fast processors to reduce the execution time of the serial fraction of a parallel computation. This paper ...
详细信息
This paper analyzes the coexistence of heterogeneity and multiprogramming in parallel systems. Using simulation models, we compare the performance of cost equivalent homogeneous and heterogeneous multiprogrammed paral...
详细信息
Summary form only given. The aims of the research are: to evaluate the suitability of existing concurrent programming systems as tools for prototyping real time engineering systems; to establish the role of formal tec...
详细信息
Summary form only given. The aims of the research are: to evaluate the suitability of existing concurrent programming systems as tools for prototyping real time engineering systems; to establish the role of formal techniques in the development of prototypes; to identify ways that this formality could subsequently be reused in the development of a well engineered software product; and to provide an environment to improve the prototyping process. Prototyping in the context of this research means the examination of areas of uncertainty in the functional requirements by execution of a software model.< >
Dynamic Base Register Caching (DBRC) [PaFa90][PaFa91] has been shown to be a useful technique for significantly reducing processor to memory address bandwidth. By caching the higher order portions of memory addresses ...
详细信息
There is a need for multicasting for different applications in all types of distributed computer systems. It is shown that multicasting can be integrated in a network as a special sublayer above the network layer. The...
详细信息
There is a need for multicasting for different applications in all types of distributed computer systems. It is shown that multicasting can be integrated in a network as a special sublayer above the network layer. The migration of the multicast and other similar services into so-called 'intelligent network' reduces delays significantly, and makes use of transmission facilities more efficient. Multicasting in point-to-point networks is examined and a minimum-weight spanning tree of connections in a network is found to be an optimal connection pattern. The distributed algorithm for the MST construction is presented as well as the mechanisms for the distribution of data packets and for synchronisation inside a multicast liaison. Future research should be focused on the introduction of switches with copying facilities into network nodes and their impact on the efficiency of multicasting.< >
暂无评论