This book investigates technologies that enable more powerful resources and improve resource utilization for end-edge-cloud networks. The authors cover tools such as federated learning (FL) and real-time inference in industrial IoT and they present a novel communication and computation integration architecture for end-edge-cloud networks. Under the considered end-edge-cloud network architecture, the authors then propose different resource scheduling schemes based on centralized and distributed deep reinforcement learning methods to improve overall resource utilization for guaranteeing the diversified quality of service (QoS) requirements from different applications. The proposed architecture and schemes can not only be adopted in future end-edge-cloud networks to efficiently manage the multi-dimensional resources in real time, but also provide useful guidelines for multi-dimensional resource scheduling scheme designing and resource utilization enhancement in complex end-edge-cloud networks with diversified data services and applications.