In this work, we analyze a distributed cooperative spectrum sensing scheme where N secondary users (SUs) of a cognitive wireless network try to agree about the primary user presence (absence) by iterative interchangin...
详细信息
In the last decade, numerous Byzantine fault-tolerant (BFT) replication protocols have been proposed in the literature. However, practically all of these solutions were designed and optimized only for certain, typical...
详细信息
ISBN:
(纸本)9783319396392;9783319396385
In the last decade, numerous Byzantine fault-tolerant (BFT) replication protocols have been proposed in the literature. However, practically all of these solutions were designed and optimized only for certain, typically very limited set of environment conditions. Despite previous efforts, no existing BFT replication protocol can guarantee stable and reasonable performance in both correct and faulty environments. In this article we attempt to address this problem by introducing Supr, a novel method for effortlessly combining multiple replication protocols into adaptive BFT solutions, which accommodate to a much wider spectrum of environment conditions than the existing BFT systems. Unlike previous approaches, Supr uses a fine-grained mechanism to monitor the parameters of the execution environment, which enables detecting and counteracting arbitrary faults exhibited in the system. To demonstrate its potential, we use Supr to create a sample BFT solution combining three existing replication protocols, each optimized for different conditions. the performed experiments demonstrate that our approach not only significantly outperforms existing solutions in varying environment conditions, but also does not introduce an observable overhead in stable environments.
the characteristic of Internet TV user behavior is quite essential for designers to optimize resource schedule and improve user experience. Withthe rapid development of Internet, both Internet TV users and STB (set t...
详细信息
Graph-based computations are used in many applications. Increasing size of analyzed data and its complexity make graph analysis a challenging task. In this paper we present performance evaluation of Java implementatio...
详细信息
ISBN:
(纸本)9783319321523;9783319321516
Graph-based computations are used in many applications. Increasing size of analyzed data and its complexity make graph analysis a challenging task. In this paper we present performance evaluation of Java implementation of Graph500 benchmark. It has been developed withthe help of the PCJ (Parallel Computations in Java) library for parallel and distributed computations in Java. PCJ is based on a PGAS (Partitioned Global Address Space) programming paradigm, where all communication details such as threads or network programming are hidden. In this paper, we present Java implementation details of first and second kernel from Graph500 benchmark. the results are compared withthe existing MPI implementations of Graph500 benchmark, showing good scalability of PCJ library.
Caching popular content at the edge of future mobile networks has been widely considered in order to alleviate the impact of the data tsunami on boththe access and backhaul networks. A number of interesting technique...
详细信息
this paper deals with sentiment analysis in text documents, especially text valence detection. the proposed solution is based on Support Vector Machines classifier. this classifier was trained with huge amount of data...
详细信息
ISBN:
(纸本)9781509012886
this paper deals with sentiment analysis in text documents, especially text valence detection. the proposed solution is based on Support Vector Machines classifier. this classifier was trained with huge amount of data and complex word combinations were analysed. For this purpose distributed learning on 112 processors was used. Datasets used for training and testing were automatically obtained from real user feedback on products from different web pages (and different product segments). the proposed solution has been evaluated with different languages - English, German, Czech and Spanish. this paper improves accuracy achieved withthe Big Data approach about 11%. the best accuracy achieved in this work was 95.31% for recognition of positive and negative text valence. the described learning is fully automatic, can be applied to any language and no complicated preprocessing is needed.
In the early phases of an automotive industry development, design time can be substantially improved by using automated tools that assist engineers to perform repetitive and time consuming tasks, speeding up automotiv...
详细信息
ISBN:
(纸本)9783319321493;9783319321486
In the early phases of an automotive industry development, design time can be substantially improved by using automated tools that assist engineers to perform repetitive and time consuming tasks, speeding up automotive products development and providing quality guarantees over otherwise error-prone processes. MERGE is such a tool from industry that prepares CAD models for electrophoretic deposition simulation. In this paper we describe the parallelization and optimization of MERGE for distributed memory parallel architectures. For this purpose we create a dynamic tree of tasks at runtime, analyze its load behavior and execute it through a master-worker compute paradigm based on different scheduling policies. Our implementation is based on a hybrid MPI-OpenMP version which results in a considerable improvement of both resource utilization and performance. Empirical performance results are presented for our new approach which achieve a speedup of up to 18 on an SMP cluster architecture.
In this paper, we investigate the cumulative distribution function (CDF) of the aggregate interference in CSMA/CA networks measured at an arbitrary time and position. We assume that nodes are deployed in an infinite t...
详细信息
Terrestrial plants have evolved remarkable adaptability that enables them to sense environmental stimuli and use this information as a basis for governing their growth orientation and root system development. In this ...
详细信息
Today the growing demand for reducing the power is not limited to household electricity saving. For businesses, it is the more important issue to effectively reduce the cost of electricity and the excess consumption u...
详细信息
ISBN:
(纸本)9783319390772;9783319390765
Today the growing demand for reducing the power is not limited to household electricity saving. For businesses, it is the more important issue to effectively reduce the cost of electricity and the excess consumption under the huge electricity. In order to achieve energy saving and energy requires, the development of energy monitoring systems to obtain information related to consumption is necessary. Accordingly, this work proposes a cloud green energy management system. Because of the data size and the computational efficiency of data analysis, we add the big data technology and cloud computing to upgrade the system performance. By building cloud infrastructure and distributed storage cluster, we adopt the open source, Hadoop, to implement the two main functions: storage and computation. Based on these two functions, the proposed system speeds up the analysis and processing of big data by using Hadoop MapReduce to access HBase. the systemic risk is thus reduced too. Both real-time data and historical data are analyzed to obtain electricity consumption behavior for real-time warning and early warning. Moreover, carbon reduction and environmental protection are also considered in the analysis. Finally, a virtualized user-interface is designed to show the proposed system functions and analysis results. the experimental results indicate the performance of the proposed system.
暂无评论