Adaptive Dynamic Virtual Network Function Placement in Mega LEO Satellite Optical Networks

Mega low earth orbit (LEO) satellite networks are becoming the key infrastructure to realize the space Internet for future 6G. The high capacity and low latency demands of next-generation satellite networks can be fulfilled by employing laser inter-satellite links (LISLs) between satellites. The computing requirements of IoT users in remote areas can be effectively served by LEO satellite optical networks. However, how to effectively utilize the limited resources of the satellite optical networks is a challenge. In this paper, a flexible service provisioning based on service function chaining (SFC) is exploited to address the challenge. The inherent issue is how to place virtual network function (VNF) instances in LEO satellite optical networks. In this paper, we first model the problem of dynamic VNF placement on satellite optical networks, with additional constraints of satellite handoff and LISLs resource allocation. Then, an adaptive-based VNF placement (A-VNFP) algorithm is proposed to address the problem. Finally, we evaluate the performance of the proposed A- VNFP algorithm for resource allocation in a mega constellation with 1584 LEO satellites.