期刊名称：TELKOMNIKA (Telecommunication Computing Electronics and Control)
印刷版ISSN：2302-9293
出版年度：2016
卷号：14
期号：4
页码：1552-1558
DOI：10.12928/telkomnika.v14i4.3606
语种：English
出版社：Universitas Ahmad Dahlan
摘要：Aiming at the waste of computing resources resulting from sequential control of running mechanism of MapReduce model on Hadoop platform，Fork/Join framework has been introduced into this model to make full use of CPU resource of each node. From the perspective of fine-grained parallel data processing, combined with Fork/Join framework，a parallel and multi-thread model，this paper optimizes MapReduce model and puts forward a MapReduce+Fork/Join programming model which is a distributed and parallel architecture combined with coarse-grained and fine-grained on Hadoop platform to Support two-tier levels of parallelism architecture both in shared and distributed memory machines. A test is made under the environment of Hadoop cluster composed of four nodes. And the experimental results prove that this model really can improve performance and efficiency of the whole system and it is not only suitable for handling tasks with data intensive but also tasks with computing intensive. it is an effective optimization and improvement to the MapReduce model of big data processing.
其他摘要：Aiming at the waste of computing resources resulting from sequential control of running mechanism of MapReduce model on Hadoop platform，Fork/Join framework has been introduced into this model to make full use of CPU resource of each node. From the perspective of fine-grained parallel data processing, combined with Fork/Join framework，a parallel and multi-thread model，this paper optimizes MapReduce model and puts forward a MapReduce+Fork/Join programming model which is a distributed and parallel architecture combined with coarse-grained and fine-grained on Hadoop platform to Support two-tier levels of parallelism architecture both in shared and distributed memory machines. A test is made under the environment of Hadoop cluster composed of four nodes. And the experimental results prove that this model really can improve performance and efficiency of the whole system and it is not only suitable for handling tasks with data intensive but also tasks with computing intensive. it is an effective optimization and improvement to the MapReduce model of big data processing.
