Migration Energy Aware Reconfigurations of Virtual Network Function Instances in NFV Architectures
Migration Energy Aware Reconfigurations of Virtual Network Function Instances in NFV Architectures
复制标题
DOI：
10.1109/access.2017.2685437
复制
发表时间：
2017-03
期刊：
影响因子：
3.9
通讯作者：
V. Eramo;M. Ammar;F. Lavacca
中科院分区：
文献类型：
--
作者：
V. Eramo;M. Ammar;F. Lavacca
网络功能虚拟化（NFV）是一种新的网络架构框架，它在共享商用服务器池上运行的软件中实现网络功能。NFV可以提供在当今不断发展的通信环境中成功竞争所需的基础设施灵活性和敏捷性。任何服务都由服务功能链（SFC）表示，SFC是要根据给定顺序执行的VNF的集合。VNF的运行需要VNF实例（VNFI）的实例化，VNF实例是在虚拟机上执行的软件模块。本文讨论了VNFI的迁移问题，需要在低流量期间关闭服务器，从而节省能源消耗。虽然合并允许节能，但它也具有负面影响，如服务质量降级或移动与要迁移的VNFI相关联的存储器所需的能量消耗。我们关注的是冷迁移，在冷迁移中，虚拟机在执行迁移之前是冗余的和挂起的。我们提出了一个迁移策略，该策略确定何时以及在何处迁移VNFI，以响应SFC请求强度的变化。目标是使由固结能和迁移能之和给出的总能耗最小。我们制定的能量感知VNFI迁移问题，并证明它是NP-难后，我们提出了一个启发式的维特比算法的基础上，能够确定具有低计算复杂度的迁移策略。通过所提出的启发式算法得到的结果表明，引入的策略允许减少迁移能量，从而降低相对于传统策略的总能耗。相对于不执行迁移的策略，能量节省可以是大约40%。
Network function virtualization (NFV) is a new network architecture framework that implements network functions in software running on a pool of shared commodity servers. NFV can provide the infrastructure flexibility and agility needed to successfully compete in today’s evolving communications landscape. Any service is represented by a service function chain (SFC) that is a set of VNFs to be executed according to a given order. The running of VNFs needs the instantiation of VNF instances (VNFIs) that are software modules executed on virtual machines. This paper deals with the migration problem of the VNFIs needed in the low traffic periods to turn OFF servers and consequently to save energy consumption. Though the consolidation allows for energy saving, it has also negative effects as the quality of service degradation or the energy consumption needed for moving the memories associated to the VNFI to be migrated. We focus on cold migration in which virtual machines are redundant and suspended before performing migration. We propose a migration policy that determines when and where to migrate VNFI in response to changes to SFC request intensity. The objective is to minimize the total energy consumption given by the sum of the consolidation and migration energies. We formulate the energy aware VNFI migration problem and after proving that it is NP-hard, we propose a heuristic based on the Viterbi algorithm able to determine the migration policy with low computational complexity. The results obtained by the proposed heuristic show how the introduced policy allows for a reduction of the migration energy and consequently lower total energy consumption with respect to the traditional policies. The energy saving can be on the order of 40% with respect to a policy in which migration is not performed.