Cost and Power Minimized Content Placement for Content Distribution Networks
With the rapid increase in Internet video services and consequent traffic, video streaming has become a crucial issue. This paper proposes minimizing the cost of a video service by optimizing the placement of content caches using a Mixed Integer Linear Programming (MILP) model and simulation. It focuses on the energy consumption and electricity cost as the objective. It explores the impact of video duration on achieved cache hit ratios by considering videos of various lengths. In addition, optimum cache locations are found and the influence of downlink demand and cache sizes on the selection of these locations is investigated. Moreover, the paper evaluates the savings in power consumption due to implementing caches at ideal locations. Furthermore, the work reveals the daily electricity cost of the video service attained under various videos lengths. Results obtained from the cache placement MILP model validated by simulation verify that maximum energy savings of 38% are achieved and electricity cost can be reduced by up to 700,000$ a day.
This work is licensed under a Creative Commons Attribution 4.0 International License.