Multi-Path Routing in Green Multi-Stage Upgrade for Bundled-Links SDN/OSPF-ECMP Networks

Abstract

This paper considers the novel problem of upgrading a legacy network into a Software Defined Network (SDN) over multiple stages and saving energy in the upgraded network or hybrid SDN. That is, in each stage, the problem at hand is to select and replace legacy switches with SDN switches, and reroute traffic to power off as many unused cables as possible to save energy. The operator has to also consider (i) the available budget at each stage, (ii) maximum path delays, (iii) maximum link utilization, (iv) per-stage increase (decrease) in traffic size (upgrade cost), and (v) the Open Shortest Path First -Equal Cost Multi-Path protocol. This paper addresses two multi-path routing scenarios: 1) non link-disjoint, and 2) linkdisjoint. It outlines a Mixed Integer Program and a heuristic algorithm for each scenario. The experimental results show that (i) both solutions for scenario-1 produce only up to 0.63% higher energy saving than those for scenario-2, (ii) the mixed integer program (heuristic algorithm) for both scenarios give an energy saving of up to 71.93% (71.64%), (iii) using a larger budget and/or number of stages increases the energy saving, and (iv) the saving achieved by the heuristic solution for each scenario is within 4% from the optimal saving.

Green Multistage Upgrade of Switches and Controllers for Bundled-Links SDNs

“…We emphasize that a solution that considers a subset of these features independently is unlikely to yield an optimal energy saving solution. Our work, similar to prior research such as [12], [24], [25], aims to maximize energy saving by switching off the maximum number of unused cables. However, reference [12] does not consider network upgrade.…”

“…In contrast, reference [21] jointly optimizes switch upgrades and controller placement under a given budget constraint over a single stage. It aims to opti- A number of works have considered network upgrade over multiple stages; e.g., [22]- [24], and [25]. The work in [22] considers switch upgrade and controller placement.…”

“…Further, the work in [23] uses shortest paths to route data traffic. On the other hand, our recent work in [24] and [25] jointly addresses switch upgrade and data traffic routing to maximize the number of unused cables. Both works consider a maximum budget at each upgrade stage, MLU, and a given delay tolerance.…”

“…Other constraints relate to the decreasing and increasing rate of upgrade cost and data traffic, respectively. Further, the work in [25] proposes the use of non-link-disjoint or link-disjoint paths. Our previous works and the work in [23], however, do not address controller placement.…”

“…However, reference [12] does not consider network upgrade. On the other hand, unlike the prior work in [17]- [20] and [23]- [25], we consider controller placement when carrying out switch upgrade and traffic routing. Different to works such as [11] and [22], our goal is to minimize energy consumption under the following constraints: 1) maximum budget that covers both the cost to upgrade a set of l-switches and deploy a set of controllers at each stage, and 2) control traffic routing via active and backup paths that are link-disjoint and satisfy a given maximum path delay and MLU threshold.…”