FedAdaSS: Federated Learning with Adaptive Parameter Server Selection Based on Elastic Cloud Resources

The rapid expansion of artificial intelligence(AI)applications has raised significant concerns about user privacy,prompting the development of privacy-preserving machine learning(ML)paradigms such as federated learning(FL).FL enables the distributed training of ML models,keeping data on local devices and thus addressing the privacy concerns of users.However,challenges arise from the heterogeneous nature of mobile client devices,partial engagement of training,and non-independent identically distributed(non-IID)data distribution,leading to performance degradation and optimization objective bias in FL training.With the development of 5G/6G networks and the integration of cloud computing edge computing resources,globally distributed cloud computing resources can be effectively utilized to optimize the FL process.Through the specific parameters of the server through the selection mechanism,it does not increase the monetary cost and reduces the network latency overhead,but also balances the objectives of communication optimization and low engagement mitigation that cannot be achieved simultaneously in a single-server framework of existing works.In this paper,we propose the FedAdaSS algorithm,an adaptive parameter server selection mechanism designed to optimize the training efficiency in each round of FL training by selecting the most appropriate server as the parameter server.Our approach leverages the flexibility of cloud resource computing power,and allows organizers to strategically select servers for data broadcasting and aggregation,thus improving training performance while maintaining cost efficiency.The FedAdaSS algorithm estimates the utility of client systems and servers and incorporates an adaptive random reshuffling strategy that selects the optimal server in each round of the training process.Theoretical analysis confirms the convergence of FedAdaSS under strong convexity and L-smooth assumptions,and comparative experiments within the FLSim framework demonstrate a reduction in training round-to-accuracy by 12%–20%compared to the Federated Averaging(FedAvg)with random reshuffling method under unique server.Furthermore,FedAdaSS effectively mitigates performance loss caused by low client engagement,reducing the loss indicator by 50%.

Rider Optimization Algorithm Based Optimal Cloud Server Selection in E-Learning

Currently,e-learning is one of the most prevalent educational methods because of its need in today’s world.Virtual classrooms and web-based learning are becoming the new method of teaching remotely.The students experience a lack of access to resources commonly the educational material.In remote loca-tions,educational institutions face significant challenges in accessing various web-based materials due to bandwidth and network infrastructure limitations.The objective of this study is to demonstrate an optimization and queueing tech-nique for allocating optimal servers and slots for users to access cloud-based e-learning applications.The proposed method provides the optimization and queue-ing algorithm for multi-server and multi-city constraints and considers where to locate the best servers.For optimal server selection,the Rider Optimization Algo-rithm(ROA)is utilized.A performance analysis based on time,memory and delay was carried out for the proposed methodology in comparison with the exist-ing techniques.The proposed Rider Optimization Algorithm is compared to Par-ticle Swarm Optimization(PSO),Genetic Algorithm(GA)and Firefly Algorithm(FFA),the proposed method is more suitable and effective because the other three algorithms drop in local optima and are only suitable for small numbers of user requests.Thus the proposed method outweighs the conventional techniques by its enhanced performance over them.

Block-based logical hierarchical cluster for distributed multimedia architecture on demand server

A structure of logical hierarchical cluster for the distributed multimedia on demand server is proposed. The architecture is mainly composed of the network topology and the resource management of all server nodes. Instead of the physical network hierarchy or the independent management hierarchy, the nodes are organized into a logically hieraxchical cluster according to the multimedia block they caches in the midderware layer. The process of a member joining/leaving or the structure adjustment cooperatively implemented by all members is concerned with decentralized maintenance of the logical cluster hierarchy. As the root of each logically hierarchical cluster is randomly mapped into the system, the logical structure of a multimedia block is dynamically expanded across some regions by the two replication policies in different load state respectively. The local load diversion is applied to fine-tune the load of nodes within a local region but belongs to different logical hierarchies. Guaranteed by the dynamic expansion of a logical structure and the load diversion of a local region, the users always select a closest idle node from the logical hierarchy under the condition of topology integration with resource management.