Currently available massively parallel supercomputers provide sufficient performance to simulate multi-dimensional, multi-variable problems in high resolution. However, the visualization of the large amounts of result...
详细信息
ISBN:
(纸本)3540422935
Currently available massively parallel supercomputers provide sufficient performance to simulate multi-dimensional, multi-variable problems in high resolution. However, the visualization of the large amounts of result data cannot be handled by traditional methods, where postprocessing modules are usually coupled to the raw data source - either by files or by data flow. Due to significant bottlenecks of the storage and communication resources, efficient techniques for data extraction and preprocessing at the source have to be developed to get a balanced, scalable system and the feasibility of a "Virtual Laboratory" scenario, where the user interacts with a multi-modal, tele-immersive virtual reality environment. In this paper we describe an efficient, distributed system approach to support three-dimensional, interactive exploration of complex results of scientific computing. Our processing chain consists of the following networked instances: 1. Creation of geometric 3D objects, such as isosurfaces, orthogonal slicers or particle sets, which illustrate the behaviour of the raw data. Our efficient visualization approach allows to handle large result data sets of simulation frameworks. It is based on processing every result data part corresponding to the domain decomposition of the parallelized simulation at the location of computation, and then collecting and exporting the generated 3D primitives. this is supported by special postprocessing routines, which provide filtering and mapping functions. 2. Storage of the generated sequence of 3D files on a separate "3D Streaming Server", which provides - controlled via "Real Time Streaming Protocol" (RTSP) - play-out capabilities for continuous 3D media streams. 3. Presentation of such 3D scene sequences as animations in a virtual reality environment. the virtual objects are embedded in a WWW page by using an advanced 3D viewer plugin, and taking advantage of high-quality rendering, stereoscopic displays and interactive navig
Large scale irregular applications involve data arrays and other data structures that are too large to fit in main memory and hence reside on disks;such applications axe called out-of-core applications. We present a d...
详细信息
ISBN:
(纸本)3540422935
Large scale irregular applications involve data arrays and other data structures that are too large to fit in main memory and hence reside on disks;such applications axe called out-of-core applications. We present a design and implementation of lip, a new parallel library that is intended to efficiently support parallel execution of irregular out-of-core codes on parallel systems. We discuss the lip structure and application interface, as well as an application that has been implemented using that interface. the library has been implemented on top of MPI/MPI-IO and, therefore, is fully portable. We introduce preliminary performance results from a template CFD codes to demonstrate the efficacy of the presented techniques.
this paper presents design and a prototype implementation of a network interface which may keep communication between processes during process migration and it may be used instead of the well-known socket interface. I...
详细信息
In this article we present an improvement on the load balancing of a parallel cluster environment, considering the MPI parallel programming paradigm and employing a mobile agent system. Our approach is to apply the mo...
详细信息
ISBN:
(纸本)3540422935
In this article we present an improvement on the load balancing of a parallel cluster environment, considering the MPI parallel programming paradigm and employing a mobile agent system. Our approach is to apply the mobile agent technology to provide a better scheduling, which could represent in a cluster configuration an enhancement on the load balancing. MPI in cluster of heterogeneous machines could lead parallel programmers to obtain frustrated results, mainly because of the lack of an even distribution of the workload in the cluster. As a result, before submitting a MPI application to a cluster, we use the Aglets mobile agent package to acquire a more precise information of machines' workload. therefore, with a more precise knowledge of the load (and characteristics) in each machine, we are ready to gather lightweight workstations to form a cluster. Our empirical results indicates that it is possible to spend less elapsed time when considering the execution of a parallel application using the agent approach in comparison to an ordinary MPI environment.
Computer simulations of granular materials are often based on the Distinct Element Method (DEM) where each grain is considered individually. Since large quantities of grains are required to perform realistic experimen...
详细信息
Message-passing programming interfaces are widely used when programming parallel applications. Systems such as PVM and MPI have been successful at providing the basic capabilities needed to implement parallel applicat...
详细信息
We discuss some new approach for derivation of computational algorithms for certain types of evolution equations. the main idea of the algorithm is to make functional transformation of variables on the base of symmetr...
详细信息
Simulation of the water flood problems often leads to solving of large sparse systems of partial differential equations. For such systems, the numerical method is very CPU-time consuming. therefore, the parallel simul...
详细信息
Peer-To-Peer systems are driving a major paradigm shift in the era of genuinely distributedcomputing. Gnutella is a good example of a Peer-To-Peer success story: a rather simple software enables Internet users to fre...
详细信息
the widespread use of World Wide Web and the diversity of Geographic Information Systems (GISs) have led to an increasing amount of research on integrating a variety of heterogeneous and autonomous GISs into a coopera...
详细信息
暂无评论