The multi-core has become inevitable and on-chip communications network to replace the traditional structure of the shared bus architecture has become the trend of the computer science development. Obverse that a gene...
详细信息
ISBN:
(纸本)9781424453320
The multi-core has become inevitable and on-chip communications network to replace the traditional structure of the shared bus architecture has become the trend of the computer science development. Obverse that a general router in mesh NoC have four access interfaces, at the border of the chip does not take full advantage of these interfaces, that is say, only router in middle of chip fully utilized. Hence, we have the main peripheral controllers are directly connected to the router of these unutilized interfaces, means border unutilized routing interface. Then the paper proposes an architecture that the NoC mesh network is divided into several virtual regions which take advantage of applications characteristics to adapt the IO external communication requirements and of the applications self internal communication patterns. Suit for parallel programming model, high I/O demanding system and utilize the existing hardware resources, our scheme is promising.
In the paper, the graph- and cloud-based tools being under development at Institute of Informatics Systems in Novosibirsk for supporting of computer science education are presented. The Web-systems Wiki GRAPP and WEGA...
详细信息
ISBN:
(数字)9783319257440
ISBN:
(纸本)9783319257440;9783319257433
In the paper, the graph- and cloud-based tools being under development at Institute of Informatics Systems in Novosibirsk for supporting of computer science education are presented. The Web-systems Wiki GRAPP and WEGA intended to help in teaching and research in graph theory, graph algorithms and their applications to computer science are described. The visual systems Higres and Visual Graph for graphs and graph algorithms are presented. The cloud system CSS aimed at supporting of functional and parallel programming is considered.
The proceedings contain 13 papers. The topics discussed include: a methodology to predict the power consumption of servers in data centres;proactive dynamic resource management in virtualized data centers;GreenSLAs fo...
ISBN:
(纸本)9781450313131
The proceedings contain 13 papers. The topics discussed include: a methodology to predict the power consumption of servers in data centres;proactive dynamic resource management in virtualized data centers;GreenSLAs for the eco-efficient management of data centres;feasibility, challenges, and performance of wireless multi-hop routing for feeder level communication in a smart grid;control and optimization meet the smart power grid: scheduling of power demands for optimal energy management;systematic classification of self-adapting algorithms for power-saving operation modes of ICT systems;energy efficient video decoding on multi-core devices;energy consumption comparison between macro-micro and public femto deployment in a plausible LTE network;EnergyPULSE: tracking sustainable behavior in office environments;and merging traffic to save energy in the enterprise.
For stream ciphers which use binary linear feedback shift registers as stream generators, fast correlation attack is one of the most important attacks. Aiming at the shortcomings of serial algorithm, the proposed tech...
详细信息
Determining the optimal configuration/ architecture for neural network could be a very time-consuming process. It includes going through big number of scenarios to simulate and validate different architecture. This pa...
详细信息
Experiments of neural network training procedure parallelization are conducted. Several styles of parallelization are described and compared, estimations of neural network size and training set size that allow speedup...
详细信息
The subsystems that compose a High Performance computing (HPC) platform (e.g. CPU, memory, storage and network) are often designed and configured to deliver exceptional performance to a wide range of workloads. As a r...
详细信息
ISBN:
(纸本)9781479984909
The subsystems that compose a High Performance computing (HPC) platform (e.g. CPU, memory, storage and network) are often designed and configured to deliver exceptional performance to a wide range of workloads. As a result, a large part of the power that these subsystems consume is dissipated as heat even when executing workloads that do not require maximum performance. Attempts to tackle this problem include technologies whereby operating systems and applications can reconfigure subsystems dynamically, such as by using Dynamic Voltage and Frequency Scaling (DVFS) for CPUs, Low Power Idle (LPI) for network components, and variable disk spinning for HDDs. Most previous work has explored these technologies individually to optimise workload execution and reduce energy consumption. We propose a framework that performs on-line analysis of an HPC system in order to identify application execution patterns without a priori information of their workload. The framework takes advantage of reoccurring patterns to reconfigure multiple subsystems dynamically and reduce overall energy consumption. Performance evaluation was carried out on Grid'5000(1) considering both traditional HPC benchmarks and real-life applications.
With the rapid development of UHV AC / DC hybrid power grid, it is required that the network Analysis Application have the ability of unified analysis and high-performance computing. In this paper, the time-consuming ...
详细信息
This paper presents the design, prototype implementation, and evaluation of a runtime management framework for structured adaptive mesh refinement applications. The framework is capable of reactively and proactively m...
详细信息
ISBN:
(数字)9783540362654
ISBN:
(纸本)3540003037
This paper presents the design, prototype implementation, and evaluation of a runtime management framework for structured adaptive mesh refinement applications. The framework is capable of reactively and proactively managing and optimizing application execution using current system and application state, predictive models for system behavior and application performance, and an agent based control network. The overall goal of this research is to enable large-scale dynamically adaptive scientific and engineering simulations on distributed, heterogeneous and dynamic execution environments such as the computational "grid".
In cloud environments, failure to fully consider the tradeoff between the energy saving and performance would lead to a sub-optimal scenario. In this paper we model the load balancing problem in cloud as a cooperative...
详细信息
ISBN:
(纸本)9781538637906
In cloud environments, failure to fully consider the tradeoff between the energy saving and performance would lead to a sub-optimal scenario. In this paper we model the load balancing problem in cloud as a cooperative game among the multiserver systems and present the structure of the Nash bargaining solution (NBS). The objective of the proposed algorithm is to minimize the energy consumption of cloud data centers while promising an acceptable average response time for their customers, in which the computing resources are provided in a greedy manner, and then the users' tasks are allocated along with the concept of the NBS. We compared the performance of the proposed algorithm with the existing approaches. Simulated results demonstrate that the proposed technique has immense potential to balance the tradeoff between the energy saving and performance for the cloud data centers.
暂无评论