As an emerging architecture, edge computing enables resource limited terminal devices to offload their computation tasks to edge servers in the vicinity, to efficiently reduce delay and energy consumption. However, the continuous expansion of network scale and rapid growth of network traffic in recent years have brought huge challenges to task offloading and resource allocation. To tackle the challenges, by integrating Knowledge Defined Networking (KDN) and edge computing technologies, we design a novel Knowledge defined Edge Computing (KEC) architecture, to achieve intelligent resource allocation and task offloading in dynamic large-scale edge computing networks. We formulate the task offloading and resource allocation optimization problem, to minimize delay and energy consumption, by considering resource requirements and controller deployment. To solve it, we present an intelligent Resource Allocation based Task Offloading (TORA) mechanism, where a Multi-Agent SD3 based resource allocation (MASD3) algorithm is devised to perform efficient resource allocation. To adapt to the rapid expansion of network scale, we design a resource Allocation based Controller Deployment and task offloading Decision (DACD) algorithm, to perform the optimal controller deployment and task offloading. Extensive simulation experiments demonstrate the effectiveness and efficiency of our proposed solution, and TORA mechanism outperforms comparison mechanisms on delay and energy consumption.

The power consumed by the Internet has been growing sharply in recent years, which accounts for a considerable proportion of the overall power consumption. To reduce high power consumption in the network, numerous proposals on power saving are proposed. Unfortunately, most of them have some side effects, such as load imbalance and even link congestion. Power saving and load balance are two conflicting optimization objectives, and the optimization solutions of them cannot be achieved at the same time. To achieve the desired tradeoff between power saving and load balance, in this article, based on software-defined networking, we first propose a Power-efficient and Load-balanced Online Flow Route framework. Then, we formulate the optimization problem of jointly taking power saving and load balance into account. After proving the problem is NP-hard, we propose the power-efficient and load-balanced flow scheduling algorithm to address it. In addition, we propose a route updating algorithm to enable path updating caused by flow scheduling. Finally, the experimental results demonstrate the proposed algorithm outperforms benchmarks on power saving, load balance, and reduces the possibility of link congestion.

With the fast development of Internet, network traffic rapidly grows and the size of Internet scale gradually becomes bigger. It brings serious challenges to Network Service Chain (NSC) deployment across multiple domains which are managed by different network operators and service providers. It is difficult for each server in the multi-domain network to obtain a global topology view and available resource information of other domains. Meanwhile, there exist various operational strategies for different network operators and service providers in the multi-domain network. How to optimally deploy NSCs in a multi-domain network to realize cost efficient and low latency service provisioning is challenging. To this end, in this paper, we study the problem of NSC deployment across multiple domains. Specifically, by leveraging Software Defined Networking (SDN) architecture, we present a novel service deployment framework to deploy Virtual Network Functions (VNFs) in the multi-domain network in a cost efficient and low latency method. Then, we establish a multi-objective optimization model with the target of minimizing end-to-end delay, service cost and operational cost with the constraints of resource allocation and VNF dependency relationships. Furthermore, we design a novel heuristic NSC deployment algorithm to solve the optimization model. In the proposed heuristic algorithm, VNF dependency relationship based NSC processing workflow optimization method and Dijkstra algorithm based NSC deployment adjustment method are presented to optimize delay and cost. Finally, extensive simulations demonstrate that the proposed heuristic deployment algorithm is efficient and outperforms comparison algorithms in terms of end-to-end delay, service cost and operational cost.