A Cost-Driven Approach to Caching-as-a-Service in Cloud-Based 5G Mobile Networks

The exploding volumes of mobile video traffic call for deploying content caches inside mobile operator networks. With in-network caching, usersu0027 requests for popular content can be served from a cache deployed at mobile gateways in vicinity of the end user. This inherently reduces the load on the content servers and the backbone of operatoru0027s network. In light of the increasing trend in virtualization of network functions, we propose a cost-effective Caching-as-a-Service (CaaS) framework for virtual video caching in 5G mobile networks. We formulate two virtual caching problems, namely maximum return on investment (MRI) and maximum offloaded traffic (MOT). MRI aims at maximizing return on caching investment by finding the best trade-off between the cost of cache storage and bandwidth savings from caching video contents in the mobile network operator (MNO)u0027s cloud. Likewise, MOT aims to maximize the offloaded traffic from MNOu0027s core and backhaul for a given budget constraint. We overcome the complexity of our problem which is formulated as a binary-integer programming (BIP) by using canonical duality theory (CDT). Experimental results obtained using the invasive weed optimization (IWO) have shown significant performance enhancement of the proposed system in terms of return on investment, quality, offloaded traffic and storage efficiency.