• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>2017 International Conference on Trends in Electronics and Informatics >Parallelization of graph labeling problem in multicore using OpenMP
【24h】

Parallelization of graph labeling problem in multicore using OpenMP

机译：使用OpenMP在多核中并行化图形标签问题

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Parallelization is the process used for reducing the time that provides an efficient result for the real time applications. Multicore architecture is a general purpose processor that consists of multiple cores on the same die and can execute programs simultaneously and parallelization can be applied effectively. Though the multiple cores are available in multicore architecture, only single core is utilized unless the programmer intervenes. It is very crucial to use the multiple cores effectively. There are many applications such as communication network addressing, X-Ray Crystallography, Radar Communication, Astronomy and Circuit design which use graph labeling problem for finding solutions. In real time, solving graph labeling problem is time consuming process when number of nodes is processed sequentially. In this work, parallelization is applied to the Graph labeling problem in multicore using OpenMP. It is found that CPU utilization is increased; speedup and execution time are reduced. After parallelization, the execution time is improved by 17.16s and CPU utilization is also improved by 50% on an average. Speedup is found to be 0.9888.
机译：并行化是用于减少时间的过程，可为实时应用程序提供有效的结果。多核体系结构是一种通用处理器，它由同一芯片上的多个内核组成，可以同时执行程序，并且可以有效地应用并行化。尽管在多核体系结构中可以使用多个核，但是除非程序员进行干预，否则只能使用单个核。有效使用多个内核非常关键。有很多应用，例如通信网络寻址，X射线晶体学，雷达通信，天文学和电路设计，这些应用都使用图形标记问题来寻找解决方案。实时地，解决图形标记问题是在顺序处理多个节点时的耗时过程。在这项工作中，使用OpenMP将并行化应用于多核中的Graph标签问题。发现增加了CPU利用率。加速和执行时间减少。并行化后，执行时间平均缩短了17.16秒，CPU利用率也平均提高了50％。发现加速为0.9888。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号