Extraction of complex data structures like vector field topologies in large-scale, unsteady flow field datasets for the interactive exploration in virtual environments cannot be carried out without parallelization str...
详细信息
Extraction of complex data structures like vector field topologies in large-scale, unsteady flow field datasets for the interactive exploration in virtual environments cannot be carried out without parallelization strategies. We present an approach based on Nested OpenMP to find critical points, which are the essential parts of velocity field topologies. We evaluate our parallelization scheme on several multi-block datasets, and present the results for various thread counts and loop schedules on all parallelization levels. Our experience suggests that upcoming massively multi-threaded processor architectures can be very advantageously for large-scale feature extractions
The n-dimensional twisted cube, denoted by TQ n , a variation of the hypercube, possesses some properties superior to the hypercube. In this paper, assuming that each vertex is incident with at least two fault-free l...
详细信息
The n-dimensional twisted cube, denoted by TQ n , a variation of the hypercube, possesses some properties superior to the hypercube. In this paper, assuming that each vertex is incident with at least two fault-free links, we show that TQ n can tolerate up to 2n - 5 edge faults, while retaining a fault-free Hamiltonian cycle. The result is optimal with respect to the number of edge faults tolerated
This paper presents vision-based strategies for decentralized stabilization of unmanned vehicle (UV) formations. The key point of the algorithms is that they only require knowledge of leader-follower relative distance...
详细信息
This paper presents vision-based strategies for decentralized stabilization of unmanned vehicle (UV) formations. The key point of the algorithms is that they only require knowledge of leader-follower relative distances or bearing angles. These data are computed using measurements from pan-controlled off-the-shelf cameras on-board following robots, eliminating sensitivity to information flow among vehicles. The approaches are based on output feedback algorithms that use high-gain observers to estimate the derivatives of the UV's relative positions. A Lyapunov stability analysis guarantees that the closed-loop system is stable and the formation error can be made arbitrarily small. A 3D virtual environment and a vision system are used to validate the proposed methodologies
There is considerable interest in designing guidance systems for UAVs that use passive sensing (such as vision), rather than active sensing which can be bulky, expensive and stealth-compromising. Here we describe an o...
详细信息
There is considerable interest in designing guidance systems for UAVs that use passive sensing (such as vision), rather than active sensing which can be bulky, expensive and stealth-compromising. Here we describe an optical sensor, based partly on principles of insect vision and optic flow analysis, for measurement and control of height above the ground. A video camera is used in conjunction with a specially shaped reflective surface to simplify the computation of optic flow, and extend the range of aircraft speeds over which accurate data can be obtained. The imaging system also provides a useful geometrical remapping of the environment, which facilitates obstacle avoidance and computation of 3-D terrain maps.
High-performance 3D graphics hardware has become as ubiquitous as floatingpoint hardware. The graphics processing unit (GPU) has, alongside the CPU, become one of the desktop's two major computational components. ...
详细信息
High-performance 3D graphics hardware has become as ubiquitous as floatingpoint hardware. The graphics processing unit (GPU) has, alongside the CPU, become one of the desktop's two major computational components. Developers design these GPUs to operate on data streams consisting of an ordered sequence of attributed primitives like vertices or fragments. Compared to conventional CPUs, GPUs consist of high-bandwidth memories and more floating-point hardware units. General-purpose computation using graphics processors (GPGP) involves the increasing use of GPUs' computational power in scientific, database, geometric, and other nongraphics applications. These GPGP applications use GPUs primarily designed for rapid rasterization of geometric primitives to shaded pixels on the screen.
We present Hyperflow, a novel visual language for information analysis that combines features from visual dataflow and visual query languages into a unified framework. Hyperflow is designed to make it easier for users...
详细信息
ISBN:
(纸本)0769524435
We present Hyperflow, a novel visual language for information analysis that combines features from visual dataflow and visual query languages into a unified framework. Hyperflow is designed to make it easier for users to retrieve, filter, and manipulate information, using databases alongside e.g. web services, in a transparent, intuitive, reproducible and traceable manner. It allows users to visually design and execute information analysis processes in a single diagram. We present Hyperflow's constructs and describe the characteristics of a prototype interface we have implemented.
In this paper, we present methodology, which enables designing and profiling macro dataflow graphs that represent computation and communication patterns for the FDTD (finite difference time domain) problem in irregul...
详细信息
In this paper, we present methodology, which enables designing and profiling macro dataflow graphs that represent computation and communication patterns for the FDTD (finite difference time domain) problem in irregular computational areas. Optimized macro dataflow graphs (MDFG) for FDTD computations are generated in three main phases: generation of initial MDFG based on wave propagation area partitioning, MDFG nodes merging with load balancing to obtain given number of macro nodes and communication optimization to minimize and balance internode data transmissions. The computation efficiency for several communication systems (MPI, RDMA RB, SHMEM) is discussed. Relations between communication optimization algorithms and overall FDTD computation efficiency are shown. Experimental results obtained by simulation are presented
By incorporating reconfigurable hardware in embedded system architectures it has become easier to satisfy the performance constraints of demanding applications while lowering system cost. In order to evaluate the perf...
详细信息
ISBN:
(纸本)0769522882
By incorporating reconfigurable hardware in embedded system architectures it has become easier to satisfy the performance constraints of demanding applications while lowering system cost. In order to evaluate the performance of a candidate architecture, the nodes (tasks) of the dataflow graphs that describe an application must be assigned to the computing resources of the architecture: programmable processors and reconfigurable FPGAs, whose run-time reconfiguration capabilities must be exploited. In this paper we present a novel design exploration tool-based on a local search algorithm with global convergence properties-which simultaneously explores choices for computing resources, assignments of nodes to these resources, task schedules on the programmable processors and context definitions for the reconfigurable circuits. The tool finds a solution that minimizes system cost while meeting the performance constraints;more precisely it lets the designer select the quality of the optimization (hence its computing time) and finds accordingly a solution with close-to-minimal cost.
As spacecraft systems evolve from simple embedded devices to become more sophisticated computing platforms with complex behaviors it is increasingly necessary to model and manage the flow of data, and to provide unifo...
详细信息
As spacecraft systems evolve from simple embedded devices to become more sophisticated computing platforms with complex behaviors it is increasingly necessary to model and manage the flow of data, and to provide uniform models for managing data that promote adaptability, yet pay heed to the physical limitations of the embedded and space environments. The mission data system (MDS) defines a software architecture in which both control theory and end-to-end data management provide the primary guiding principles. This paper describes how the MDS architecture facilitates data accountability and storage resource management.
The ability to dynamically change data input to a computation is a key feature enabling simulation to be used in many applications. In this study, computation of geophysical mass flow is updated on the fly by changing...
详细信息
The ability to dynamically change data input to a computation is a key feature enabling simulation to be used in many applications. In this study, computation of geophysical mass flow is updated on the fly by changing terrain data. Accommodating such changes in a parallel environment entails new developments in parallel data management and gridding. Adaptivity, and in particular unrefinement, is critical for maintaining parallel efficiency. The application under study in this work is the result of a multidisciplinary collaboration between engineers, mathematicians, geologists, and hazard assessment personnel. In addition, adaptive gridding enables efficient use of computational resources, allowing for run-time determination of optimal computing resources. Combining these attributes allows run time conditions to inform calculations, which in turn provide up-to-date information to hazard management personnel.
暂无评论