This paper presents the design and the implementation of a compiler and runtime infrastructure for automatic program distribution. We are building a research infrastructure that enables experimentation with various pr...
详细信息
ISBN:
(纸本)0769523129
This paper presents the design and the implementation of a compiler and runtime infrastructure for automatic program distribution. We are building a research infrastructure that enables experimentation with various program partitioning and mapping strategies and the study of automatic distribution' s effect on resource consumption (e.g., CPU, memory, communication). Since many optimization techniques are faced with conflicting optimization targets (e.g., memory and communication), we believe that it is important to be able to study their interaction. We present a set of techniques that enable flexible resource modeling and program distribution. These are: dependence analysis, weighted graph partitioning, code and communication generation, and profiling. We have developed these ideas in the context of the Java language. We present in detail the design and implementation of each of the techniques as part of our compiler and runtime infrastructure. Then, we evaluate our design and present preliminary experimental data for each component, as well as for the entire system.
We propose a new random beamforming technique in forward link multiuser MIMO systems that exploits efficient multiuser diversity and spatial multiplexing gain. The common channels are allocated to more than one user s...
详细信息
This session relates directly to this year's conference theme Sparking Synergies: Bringing Research and Practice Together as it brings together researchers from different disciplines (information science and educa...
This session relates directly to this year's conference theme Sparking Synergies: Bringing Research and Practice Together as it brings together researchers from different disciplines (information science and educational technology) to discuss current issues and trends related to Web usage across student populations. “Complex design problems require more knowledge than any single person possesses” (Arias, Eden, Fischer, Gorman, and Scharff, 2000, p. 84), thus each panelist will contribute highlights of his or her current research regarding the design, information, and pedagogical, needs of specific student populations (K-12 to higher education). After the panelists have presented their findings Diane Nahl will intertwine the threads of the panelists' conversations and, with the audience's assistance, begin to weave a fabric to uncover common usability issues.
Determination of an isothetic polygonal approximation of the outer or inner contour of an object is a challenging problem with numerous applications to pattern recognition and image processing. In this paper, a novel ...
详细信息
A new deterministic broadcast protocol for an ad hoc network is proposed in this paper which avoids re-computation of the transmission schedule, even when the topology of the network changes due to the mobility of the...
详细信息
We propose a new random beamforming technique in forward link multiuser MIMO systems that exploits efficient multiuser diversity and spatial multiplexing gain. The common channels are allocated to more than one user s...
详细信息
We propose a new random beamforming technique in forward link multiuser MIMO systems that exploits efficient multiuser diversity and spatial multiplexing gain. The common channels are allocated to more than one user simultaneously and multiple antennas at the receivers are used for interference nulling, resulting in higher throughput and reduced average waiting time. We show that the throughput of the proposed scheme outperforms the upper bound of the conventional scheme.
Implicit relevance feedback (IRF) is the process by which a search system unobtrusively gathers evidence on searcher interests from their interaction with the system. IRF is a new method of gathering information on us...
详细信息
This paper presents the design and the implementation of a compiler and runtime infrastructure for automatic program distribution. We are building a research infrastructure that enables experimentation with various pr...
详细信息
This paper presents the design and the implementation of a compiler and runtime infrastructure for automatic program distribution. We are building a research infrastructure that enables experimentation with various program partitioning and mapping strategies and the study of automatic distribution's effect on resource consumption (e.g., CPU, memory, communication). Since many optimization techniques are faced with conflicting optimization targets (e.g., memory and communication), we believe that it is important to be able to study their interaction. We present a set of techniques that enable flexible resource modeling and program distribution. These are: dependence analysis, weighted graph partitioning, code and communication generation, and profiling. We have developed these ideas in the context of the Java language. We present in detail the design and implementation of each of the techniques as part of our compiler and runtime infrastructure. Then, we evaluate our design and present preliminary experimental data for each component, as well as for the entire system.
Outliers are data values that lie away from the general cluster of other data values. Detecting the outliers of a dataset is an important research topic for data cleaning and finding new useful knowledge in many resea...
详细信息
A hybrid atomistic-continuum simulation approach has been implemented to study strain relaxation in lattice-mismatched Si∕Si3N4 nanopixels on a Si(111) substrate. We couple the molecular-dynamics (MD) and finite-elem...
A hybrid atomistic-continuum simulation approach has been implemented to study strain relaxation in lattice-mismatched Si∕Si3N4 nanopixels on a Si(111) substrate. We couple the molecular-dynamics (MD) and finite-element simulation approaches to provide an atomistic description near the interface and a continuum description deep into the substrate, increasing the accessible length scales and greatly reducing the computational cost. The results of the hybrid simulation are validated against full multimillion-atom MD simulations. We find that strain relaxation in Si∕Si3N4 nanopixels may occur through the formation of a network of interfacial domain boundaries reminiscent of interfacial misfit dislocations. They result from the nucleation of domains of different interfacial bonding at the free edges and corners of the nanopixel, and subsequent to their creation they propagate inwards. We follow the motion of the domain boundaries and estimate a propagation speed of about ∼2.5×103m∕s. The effects of temperature, nanopixel architecture, and film structure on strain relaxation are also investigated. We find: (i) elevated temperature increases the interfacial domain nucleation rates; (ii) a thin compliant Si layer between the film and the substrate plays a beneficial role in partially suppressing strain relaxation; and (iii) additional control over the interface morphology may be achieved by varying the film structure.
暂无评论