Edge Provisioning with Flexible Server Placement

Abstract

We present <inline-formula><tex-math notation="LaTeX">$\sf {Tentacle}$</tex-math></inline-formula> , a decision support framework to provision edge servers for online services providers (OSPs). <inline-formula><tex-math notation="LaTeX">$\sf {Tentacle}$</tex-math> </inline-formula> takes advantage of the increasingly flexible edge server placement, which is enabled by new technologies such as edge computing platforms, cloudlets and network function virtualization, to optimize the overall performance and cost of edge infrastructures. The key difference between <inline-formula><tex-math notation="LaTeX"> $\sf {Tentacle}$</tex-math> </inline-formula> and traditional server placement approaches lies on that <inline-formula><tex-math notation="LaTeX"> $\sf {Tentacle}$</tex-math> </inline-formula> can discover proper unforeseen edge locations which significantly improve the efficiency and reduce the cost of edge provisioning. We show how <inline-formula><tex-math notation="LaTeX">$\sf {Tentacle}$</tex-math> </inline-formula> effectively identifies promising edge locations which are close to a collection of users merely with inaccurate network distance estimation methods, e.g., geographic coordinate (GC) and network coordinate systems (NC). We also show how <inline-formula><tex-math notation="LaTeX">$\sf {Tentacle}$</tex-math></inline-formula> comprehensively considers various pragmatic concerns in edge provisioning, such as traffic limits by law or ISP policy, edge site deployment and resource usage cost, over-provisioning for fault tolerance, etc., with a simple optimization model. We simulate <inline-formula><tex-math notation="LaTeX">$\sf {Tentacle}$</tex-math></inline-formula> using real network data at global and county-wide scales. Measurement-driven simulations show that with a given cost budget <inline-formula> <tex-math notation="LaTeX">$\sf {Tentacle}$</tex-math></inline-formula> can improve user performance by around 10-45 percent at global scale networks and 15-35 percent at a country-wide scale network.