One of the important ultimatums in enhancing the fifth generation (5G) network's capacity is the performance limit due to spectrum occupancy. The mmWave technology overlay on 5G heterogeneous network (HetNet) and caching the contents are proposed as the solutions to this problem. By reducing the backhaul links occupancy, increasing the access link utilization and caching, the impact of spectrum occupancy problem can be minimized. In our earlier work, differential Quality of Service (QoS) was provisioned by managing the cache and backhaul resources using machine learning techniques. In this work, an integrated solution combining content, cache and user management is proposed to maximize network utilization and QoS. The problem of effective utilization of the network at the same time ensuring the QoS for users are solved as a multi objective optimization problem with the aid of hybrid meta heuristics with complementary exploration and exploitation capability. The user association to base station is made adaptive to load and cache hit ratio at the base station. By increasing the content proximity to the users, the load of backhaul links is minimized. Through these integrated management strategies, the proposed solution is able to provide higher QoS compared to existing works in terms of reduction in packet drop by 6%, reduction in delay by 39%, increase in network throughput by 8% and a consistent cache hit ratio more than 85%.INDEX TERMS MmWave backhaul; hetnet; caching; multi criteria optimization, hybrid meta-heuristics.