Energy consumption for datacenter has grown significantly and the trend is still growing due to the increasing\npopularity of cloud computing. Datacenter networks (DCNs), however, are starting to consume a greater portion\nof overall energy in comparison to servers used in datacenters due to advanced virtualization techniques. On the\nother hand, devices in a DCN often remain under-utilized. There are various DCN architectures. This paper proposes an\napproach called Green Spine Switch Management System (GSSMS) for Spine-Leaf topology based DCNs. The objective\nof the approach is to reduce energy consumption used by the network for a Spine-Leaf topology-based\ndatacenter. The primary idea of GSSMS is to monitor the dynamic workload and only keep Spine switches that\nare necessary for handling the current network traffic. We have developed an adaptive management system to\ncontrol the number of Spine switches in a Spine-Leaf DCN for efficient energy consumption. Further, we have\nperformed extensive simulation using CloudSim for a number of scenarios. The simulation results demonstrate\nthat our proposed GSSMS can effectively save energy by as much as 63 % of the energy consumed by a\ndatacenter comprising a fixed static set of Spine switches.
Loading....