An efficient load balancing approach for service function chain mapping

Edge intelligence for service function chain deployment in NFV-enabled networks

“…Another research [11], proposed a solution for balancing the load among VNFs by making a tree of SCs and then developing a flow-hash technique to distribute load among service chains. Finally, authors in [12], proposed an offline VNF placement algorithm which aims to jointly balance the network and server load. They leveraged the water filling algorithm, for this purpose.…”

“…The constraints defined in Eqs. (11) and (12), enforce that each service flow uses only one path to carry its traffic, and all VNFs of a given service chain will be placed only on the selected path, respectively.…”

“…However, despite the importance of the subject, less effort has been made about jointly balancing the network and server load. In addition, the existing load balancing approaches in the literature [11,12], have focused on the offline model-based heuristics or using optimization solvers that are limited to particular system models assumed by authors and they are not applicable for different scenarios or other real-time networks. Moreover, these models need prior information about the network traffic which is not available in practice.…”