parallel and distributed systems engineers are always looking for a way to speed-up their programs. They sometimes forget that well-structured programs are more flexible, and therefore easier to modify or restructure ...
详细信息
parallel and distributed systems engineers are always looking for a way to speed-up their programs. They sometimes forget that well-structured programs are more flexible, and therefore easier to modify or restructure in order to improve performance or to map onto a particular architecture. This paper illustrates a systematic way of designing well-structured parallel and distributed programs. The method is based on SASD, one of the most popular methods for the analysis and design of sequential systems, and CSP, a formalism for specifying the behaviour of communicating systems. The influence of SASD is evident in the way diagrams are used during the various phases of the development. CSP allows us to formally verify and transform the programs. The main feature of our method is the ability to reuse behavioural specifications, the way the components synchronise and communicate, and provide rules to verify and transform the design structure.< >
Physical modeling synthesis is a powerful means of access to a wide variety of synthetic sounds of an acoustic character-one longstanding design principle underlying such methods has been, and continues to be modulari...
详细信息
Building distributed multimedia applications from pre-existing, reusable components will only take place when multimedia application developers can easily understand, extend and modify software components according to...
详细信息
作者:
Hamma, B.Parallel Algorithms Team
CERFACS European Centre for Research Advanced Training in Scientific Computing 42 av. Coriolis Toulouse-Cedex F-31057 France
This paper describes the software developed at CNES within the context of the VESTA Franco-Soviet program, for the optimization of interplanetary trajectories. The sequentiel software, intended for the determination o...
详细信息
Exploration using borehole drilling is a key activity in determining the most appropriate locations for the petroleum industry to develop oil fields. However estimating the amount of Oil In Place (OIP) relies on compu...
详细信息
Exploration using borehole drilling is a key activity in determining the most appropriate locations for the petroleum industry to develop oil fields. However estimating the amount of Oil In Place (OIP) relies on computing with a very significant number of geological models which, due to the ever increasing capability to capture and refine data, is becoming infeasible. As such data reduction techniques are required to reduce this set down to a much smaller yet still fully representative ensemble. In this paper we explore different approaches to identifying the key grouping of models, based on their most important features, and then using this information select a reduced set which we can be confident fully represent the overall model space. The result of this work is an approach which enables us to describe the entire state space using only 0.5% of the models, along with a series of lessons learnt. The techniques that we describe are not only applicable to oil and gas exploration, but also more generally to the HPC community as we are forced to work with reduced data-sets due to the rapid increase in data collection capability.
Increasing interest is being shown in the use of Java for scientific applications. The Java Grande benchmark suite was designed with such applications primarily in mind. The perceived lack of performance of Java still...
详细信息
ISBN:
(纸本)1581133596
Increasing interest is being shown in the use of Java for scientific applications. The Java Grande benchmark suite was designed with such applications primarily in mind. The perceived lack of performance of Java still deters many potential users, despite recent advances in just-in-time (JIT) and adaptive compilers. There are however few benchmark results available comparing Java to more traditional languages such as C and Fortran. To address this issue, a sub-set of the Java Grande Benchmarks have been re-written in C and Fortran allowing direct performance comparisons between the three languages. The performance of a range of Java execution environments, C and Fortran compilers have been tested across a number of platforms using the suite. These demonstrate that on some platforms (notably Intel Pentium) the performance gap is now quite small.
This project investigated the feasibility of building a General Purpose Architecture Simulator on an MIMD transputer network. In the course of this study an Occam2 simulation of the 88100 Reduced Instruction Set micro...
详细信息
This paper presents the development of the cloud software services for computational analysis of blood flows on a private university cloud. The main focus is on the software service level built on the top of the compu...
详细信息
Distributed Virtual Environments (DVEs), such as online games, military simulations, collaborative design, etc., are very popular nowadays. To support large-scale DVEs, a multi-server architecture is usually employed,...
详细信息
Distributed Virtual Environments (DVEs), such as online games, military simulations, collaborative design, etc., are very popular nowadays. To support large-scale DVEs, a multi-server architecture is usually employed, and the virtual world is partitioned into multiple zones for load distribution. The client assignment problem arises when assigning the participating clients to the servers. Current approaches usually assign clients to servers according to the locations of clients in the virtual world, i.e., clients interacting in the same zone of the virtual world will be assigned to the same server. This approach may degrade the interactivity of DVEs if the network delay from a client to its assigned server is large. In this paper, we formulate the client assignment problem, and propose two algorithms to assign clients to servers in a more efficient way. The proposed algorithms are based on the heuristics developed for the well-known Terminal Assignment problem. Simulation results with the BRITE Internet Topology Generator show that our algorithms are effective in enhancing the interactivity of DVEs.
暂无评论