• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>IEEE International Conference on Communication Technology >Towards Performance Optimization of Network Service Chains with Multi-Ingress and Single-Egress
【24h】

Towards Performance Optimization of Network Service Chains with Multi-Ingress and Single-Egress

机译：利用多进入和单次出口的网络服务链的性能优化

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The emerging and developing of Service Function Chain (SFC) considerably promote the performance of network services, especially as various efficient SFC deployment methods are proposed and adopted. However, current SFC deployment methods mainly focus on single-ingress and single-egress scenarios by arranging Virtual Machines (VMs) properly, ignoring a much more popular application pattern that service requests are imported from different ingresses and guided to a common egress. To improve the performance of network service chains in such multi-ingress and single-egress context, this paper presents an Optimized VMs Placement Model on multi-ingress and single-egress Network (OVPMN) for lower delay and energy consumption. OVPMN encloses the formulation of Integer Linear Program (ILP) and models the multi-ingress problem as a convex optimization problem. A mutated K-shortest path algorithm is proposed to minimize the combination of total delay and energy consumptions by introducing conditional probability. Numerical simulations are implemented to evaluate the proposed algorithm, and the results show that OVPMN has better performance than traditional SFC schemes.
机译：服务功能链（SFC）的新兴和发展大大促进了网络服务的性能，特别是提出和采用了各种有效的SFC部署方法。然而，目前证监会的部署方法主要是通过安排虚拟机（VM）正常，忽略一个更受欢迎的应用程序模式的服务请求都从不同进口入节点，并引导到共同的出口集中在单入口和单流出情形。为了提高网络服务链的性能，在这种多进入和单个出口背景下，本文介绍了用于较低延迟和能量消耗的多进入和单出口网络（OVPMN）上的优化VMS放置模型。 OVPMN包含整数线性程序（ILP）的配方，并将多进入问题模拟为凸优化问题。提出了一种突变的k最短路径算法，以通过引入条件概率来最小化总延迟和能量消耗的组合。实现了数值模拟以评估所提出的算法，结果表明，OVPMN具有比传统SFC方案更好的性能。

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