TY - JOUR

T1 - VNF Orchestration and Power-Disjoint Traffic Flow Routing for Optimal Communication Robustness in Smart Grid With Cyber-Physical Interdependence

AU - Kong, Peng Yong

AU - Jiang, Yuming

N1 - Publisher Copyright:
© 2004-2012 IEEE.

PY - 2022/12/1

Y1 - 2022/12/1

N2 - We explore the use of software-defined networking (SDN) technology in building a communication network for smart grid. With cyber-physical interdependence, such communication network may suffer from cross-network cascading failures. To prevent the failures, we perform virtual network function (VNF) orchestration jointly with power-disjoint routing. Our work is novel in proposing an efficient scheme to find power-disjoint routes at the same time of performing VNF orchestration. We formulate an optimization to maximize the ratio of power-disjoint route count to VNF orchestration cost. The optimization has a non-linear non-convex objective function. We propose a two-level hierarchical solution approach. At higher level, the scheme converts the problem into a fractional maximum flow circulation, which can be solved using simplex method to find the maximum number of power-disjoint routes. Given a higher level solution, the lower level aims to minimize the VNF orchestration cost while satisfying VNF chaining and placement requirements. This lower level hierarchy uses the Dijkstra's algorithm in building a sequence of minimum spanning trees, each roots at the current VNF hosting node in a VNF chain. Extensive simulation results confirm that the proposed scheme can find the maximum number of power-disjoint routes and minimize the cost within a second, for a system with 120 communication nodes. The results show that the number of power-disjoint routes can be increased by increasing either the number of nodes or node degree, but only the node degree can keep the cost flat. Therefore, one should build a robust software-defined smart grid communication network by enhancing node connectivity.

AB - We explore the use of software-defined networking (SDN) technology in building a communication network for smart grid. With cyber-physical interdependence, such communication network may suffer from cross-network cascading failures. To prevent the failures, we perform virtual network function (VNF) orchestration jointly with power-disjoint routing. Our work is novel in proposing an efficient scheme to find power-disjoint routes at the same time of performing VNF orchestration. We formulate an optimization to maximize the ratio of power-disjoint route count to VNF orchestration cost. The optimization has a non-linear non-convex objective function. We propose a two-level hierarchical solution approach. At higher level, the scheme converts the problem into a fractional maximum flow circulation, which can be solved using simplex method to find the maximum number of power-disjoint routes. Given a higher level solution, the lower level aims to minimize the VNF orchestration cost while satisfying VNF chaining and placement requirements. This lower level hierarchy uses the Dijkstra's algorithm in building a sequence of minimum spanning trees, each roots at the current VNF hosting node in a VNF chain. Extensive simulation results confirm that the proposed scheme can find the maximum number of power-disjoint routes and minimize the cost within a second, for a system with 120 communication nodes. The results show that the number of power-disjoint routes can be increased by increasing either the number of nodes or node degree, but only the node degree can keep the cost flat. Therefore, one should build a robust software-defined smart grid communication network by enhancing node connectivity.

KW - cyber-physical interdependence

KW - network function virtualization

KW - power-disjoint route

KW - Smart grid

KW - software-defined networking

UR - http://www.scopus.com/inward/record.url?scp=85127768986&partnerID=8YFLogxK

U2 - 10.1109/TNSM.2022.3165219

DO - 10.1109/TNSM.2022.3165219

M3 - Article

AN - SCOPUS:85127768986

SN - 1932-4537

VL - 19

SP - 4479

EP - 4490

JO - IEEE Transactions on Network and Service Management

JF - IEEE Transactions on Network and Service Management

IS - 4

ER -