With the rapid proliferation of big data, real-time processing of huge datasets becomes a challenging task;primarily because of their heterogeneous nature. Due to this, one of the most serious concerns of the modern c...
详细信息
ISBN:
(纸本)9781538631805
With the rapid proliferation of big data, real-time processing of huge datasets becomes a challenging task;primarily because of their heterogeneous nature. Due to this, one of the most serious concerns of the modern cloud data centers is massive energy consumption during job execution. Hence, energy-aware task scheduling with data placement are considered as two important parameters for enhanced energy efficiency of modern cloud data centers. Moreover, considering the "pay-per-use" model of cloud computing infrastructure, it is important to maintain desirable service level agreement (SLA) while attaining improved data locality. Poor task scheduling decisions with limited focus of data locality are the prime reasons for escalated data communications and energy utilization levels. In order to deal with the aforementioned issues, data locality-aware energy-efficient (EnLoc) scheme for task scheduling and data placement has been proposed, particularly for MapReduce framework. The proposed EnLoc scheme is a multi-objective optimization problem (MOOP) and is solved using multi-objectiveevolutionaryalgorithm with "Tchebycheff decomposition";wherein the formulated MOOP is decomposed into theoretically finite number of subproblems to get optimal scheduling and placement decisions. The proposed scheme has been evaluated on real-time data traces acquired from OpenCloud Hadoop Cluster. The results obtained clearly demonstrate that the proposed EnLoc scheme outperforms the existing schemes in terms of energy efficiency, SLA assurance, and data locality.
In order to achieve robust performance of preserving significant image details while removing noise for image segmentation, this paper presents a multi-objectiveevolutionary fuzzy clustering (MOEFC) algorithm to conv...
详细信息
In order to achieve robust performance of preserving significant image details while removing noise for image segmentation, this paper presents a multi-objectiveevolutionary fuzzy clustering (MOEFC) algorithm to convert fuzzy clustering problems for image segmentation into multi-objective problems. The multi-objective problems are optimized by multi-objective evolutionary algorithm with decomposition. The decomposition strategy is adopted to project the multi-objective problem into a number of subproblems. Each sub-problem represents a fuzzy clustering problem incorporating local information for image segmentation. Opposition-based learning is utilized to improve search capability of the proposed algorithm. Two problem-specific techniques, an adaptive weighted fuzzy factor and a mixed population initialization, are introduced to improve the performance of the algorithm. Experiment results on synthetic and real images illustrate that the proposed algorithm can achieve a trade-off between preserving image details and removing noise for image segmentation. (C) 2016 Elsevier B.V. All rights reserved.
暂无评论