Serverless edge computing, characterized by fine-grained resource allocation and rapid task scheduling, is effectively implemented in edge clouds to support a diverse array of Deep Neural Network (DNN) based inference tasks. However, before executing inference tasks, the system needs to load DNN models into a container, a process known as cold start. The cold start introduces significant latency, thereby prolonging the task completion time. In particular, model loading time constitutes approximately 50%–70% of the overall task lifecycle, making it comparable to the duration of inference execution. Fortunately, we notice that some layers are required by multiple models and only need to be loaded once if these tasks are scheduled onto one server, i.e., DNN layer sharing. However, the computational resource differences among edge servers and their limited memory capacity make task scheduling and layer loading decisions particularly challenging. Therefore, to fully leverage the potential of DNN layer sharing, we investigate the Layer Sharing aware Task Scheduling (LSTS) problem with the goal of minimizing the task completion time. We formulate it into a Quadratic Integer Programming (QIP) problem and linearize it into an Integer Linear Programming (ILP) form, which is then proved as NP-hard. To tackle the computation complexity, we propose a Randomized Rounding-based Layer-Sharing-Aware Task Scheduling algorithm (LSTS-RR). Through comprehensive experimental evaluations, we confirm the effectiveness of our algorithm, as it reduces task completion time by more than 36% compared to other state-of-the-art approaches across a range of widely recognized DNN models.

Summary Cloud computing (CC) provides dynamic hiring of server abilities as scalable virtualized services to end‐users. However, data center hosting wastes massive amounts of energy resulting in high operational costs and carbon footprints. Also, virtualization is one of CC's main features, and physical resources are delivered by virtual machine (VM). Therefore, in the present article, a new method is provided to improve the VM energy consumption and execution time in the VM migration problem using a hybrid optimization algorithm. Since this issue is one of the famous NP‐hard problems, a method is proposed in this article works based on genetic algorithm (GA) and particle swarm optimization (PSO) algorithm. The hybrid algorithm uses a GA to dominate PSO algorithms' constraints, such as weak convergence and stymie in global optima. The CloudSim simulator is employed to show the efficiency of the method compared to others. Using this method will keep the proficiency and power performance of the data centers at the same level. The results showed that energy consumption in the proposed method is better than the other three methods and has been improved by an average of 23.19%. Also, the results showed that execution time is better than the other three methods and has been improved by an average of 29.01%.

The following topics are dealt with: learning (artificial intelligence); optimisation; telecommunication traffic; Internet; cloud computing; computational complexity; mobile computing; resource allocation; security of data; and telecommunication network routing.

The function of digital library is not limited to information management now, but also has more relationship with community. P2P networks are highly scalable platforms, which should be suitable for personal digital library, but the contradiction between efficiency and autonomy holds back their progress in applying them to DL. In this paper, we propose an autonomous flexible content addressable network (Flexible CAN), in which nodes can change their position freel. Flexible CAN has the advantages of both efficiency and autonomy. It can be used not only for information managing and retrieving, but also autonomic detect group of users with the same interest. Thus the group becomes a community, and users can share their resources and experiences conveniently.