This paper focuses on the co-design of networked gossip protocol and distributed control for interconnected large-scale systems to achieve closed-loop stability and networked system's H ∞ performance. In the proposed distributed system, each subcontroller relies on locally estimated states as well as the states shared by its scheduled neighbours through the networked gossip protocol. The communication probabilities in gossip protocol are co-designed with the distributed controllers for better system performance. By utilising the properties of Lyapunov stability and matrix theory, sufficient conditions are developed to ensure the stability and H ∞ performance for the closed-loop system in the mean-square sense. However, the obtained sufficient conditions include nonlinear terms and some equality constraints which are difficult to be solved. To address the non-convex problem, this paper develops a novel solving algorithm which incorporates penalty-based genetic algorithm and linear matrix inequalities (pGA-LMI algorithm). By the proposed pGA-LMI algorithm, the distributed controllers and the communication probabilities in gossip protocol are co-designed, and the H ∞ performance can be optimised simultaneously. Finally, the feasibility and effectiveness of the proposed co-design method are ensured by applying it to some numerical examples and a four-area interconnected power system.

Decentralized federated learning (D-FL) allows clients to aggregate learning models locally, offering flexibility and scalability. Existing D-FL methods use gossip protocols, which are inefficient when not all nodes in the network are D-FL clients. This article puts forth a new D-FL strategy, termed route-and-aggregate (R&A) D-FL, where participating clients exchange models with their peers through established routes (as opposed to flooding) and adaptively normalize their aggregation coefficients to compensate for communication errors. The impact of routing and imperfect links on the convergence of R&A D-FL is analyzed, revealing that convergence is minimized when routes with the minimum end-to-end (E2E) packet error rates (PERs) are employed to deliver models. Our analysis is experimentally validated through three image classification tasks and two next-word prediction tasks, utilizing widely recognized datasets and models. R&A D-FL outperforms the flooding-based D-FL method in terms of training accuracy by 35% in our tested ten-client network, and shows strong synergy between D-FL and networking. In another test with ten D-FL clients, the training accuracy of R&A D-FL with communication errors approaches that of the ideal centralized federated learning (C-FL) without communication errors, as the number of routing nodes (i.e., nodes that do not participate in the training of D-FL) rises to 28.

Federated Learning (FL) presents a promising avenue for collaborative model training among medical centers, facilitating knowledge exchange without compromising data privacy. However, vanilla FL is prone to server failures and rarely achieves optimal performance on all participating sites due to heterogeneous data distributions among them. To overcome these challenges, we propose Gossip Contrastive Mutual Learning (GCML), a unified framework to optimize personalized models in a decentralized environment, where Gossip Protocol is employed for flexible and robust peer-to-peer communication. To make efficient and reliable knowledge exchange in each communication without the global knowledge across all the sites, we introduce deep contrast mutual learning (DCML), a simple yet effective scheme to encourage knowledge transfer between the incoming and local models through collaborative training on local data. By integrating DCML with other efforts to optimize site-specific models by leveraging useful information from peers, we evaluated the performance and efficiency of the proposed method on three publicly available datasets with different segmentation tasks. Our extensive experimental results show that the proposed GCML framework outperformed both centralized and decentralized FL methods with significantly reduced communication overhead, indicating its potential for real-world deployment. Upon the acceptance of manuscript, the code will be available at: https://github.com/ CUMC-Yuan-Lab/GCML.

This research paper explores the implementation of gossip protocols in cloud native framework through network modeling and simulation analysis. Gossip protocol is known for their decentralized and fault-tolerant nature. Simulating gossip protocols with conventional tools may face limitations in flexibility and scalability, complicating analysis, especially for larger or more diverse networks. In this paper, gossip protocols are tested within the context of cloud native computing, which leverages its scalability, flexibility, and observability. The study aims to assess the performance and feasibility of gossip protocols within cloud-native settings through a simulated environment. The paper delves into the theoretical foundation of gossip protocol, highlights the core components of cloud native computing, and explains the methodology employed in the simulation. A detailed guide has been provided on utilizing cloud-native frameworks to simulate gossip protocols across varied network environments. The simulation analysis provides insights into gossip protocols' behavior in distributed cloud-native systems, evaluating aspects of scalability, reliability, and observability. This investigation contributes to understanding the practical implications and potential applications of gossip protocol within modern cloud-native architectures, which can also apply to conventional network infrastructure.

This research paper explores the implementation of gossip protocols in cloud native framework through network modeling and simulation analysis. Gossip protocol is known for their decentralized and fault-tolerant nature. Simulating gossip protocols with conventional tools may face limitations in flexibility and scalability, complicating analysis, especially for larger or more diverse networks. In this paper, gossip protocols are tested within the context of cloud native computing, which leverages its scalability, flexibility, and observability. The study aims to assess the performance and feasibility of gossip protocols within cloud-native settings through a simulated environment. The paper delves into the theoretical foundation of gossip protocol, highlights the core components of cloud native computing, and explains the methodology employed in the simulation. A detailed guide has been provided on utilizing cloud-native frameworks to simulate gossip protocols across varied network environments. The simulation analysis provides insights into gossip protocols’ behavior in distributed cloud-native systems, evaluating aspects of scalability, reliability, and observability. This investigation contributes to understanding the practical implications and potential applications of gossip protocol within modern cloud-native architectures, which can also apply to conventional network infrastructure.

Leaderless consensus algorithms represent a significant advancement in distributed systems, eliminating single points of failure while enhancing fault tolerance. However, these systems face considerable communication overhead challenges as they scale to include numerous nodes across global networks. This article examines techniques that reduce message traffic while maintaining effective consensus, including quorum-based voting, gossip protocols, message aggregation and compression, asynchronous communication, and partial synchrony approaches. These methods deliver substantial benefits such as improved scalability, reduced latency, lower resource requirements, and enhanced fault tolerance. Despite these advantages, implementation presents several challenges, including consistency-efficiency trade-offs, complex implementations, security vulnerabilities, and parameter tuning difficulties. Looking forward, emerging innovations such as adaptive protocols, network-aware optimizations, hardware acceleration, hybrid approaches, and privacy-preserving techniques promise to further revolutionize communication efficiency in distributed systems.

Internet-of-Things (IoT) devices are interconnected objects embedded with sensors and software, enabling data collection and exchange. These devices encompass a wide range of applications, from household appliances to industrial systems, designed to enhance connectivity and automation. In distributed IoT networks, achieving reliable decision-making necessitates robust consensus mechanisms that allow devices to agree on a shared state of truth without reliance on central authorities. Such mechanisms are critical for ensuring system resilience under diverse operational conditions. Recent research has identified three common limitations in existing consensus mechanisms for IoT environments: dependence on synchronised networks and clocks, reliance on centralised coordinators, and suboptimal performance. To address these challenges, this paper introduces a novel consensus mechanism called Randomised Asynchronous Consensus with Efficient Real-time Sampling (RACER). The RACER framework eliminates the need for synchronised networks and clocks by implementing the Sequenced Probabilistic Double Echo (SPDE) algorithm, which operates asynchronously without timing assumptions. Furthermore, to mitigate the reliance on centralised coordinators, RACER leverages the SPDE gossip protocol, which inherently requires no leaders, combined with a lightweight transaction ordering mechanism optimised for IoT sensor networks. Rather than using a blockchain for transaction ordering, we opted for an eventually consistent transaction ordering mechanism to specifically deal with high churn, asynchronous networks and to allow devices to independently and deterministically order transactions. To enhance the throughput of IoT networks, this paper also proposes a complementary algorithm, Peer-assisted Latency-Aware Traffic Optimisation (PLATO), designed to maximise efficiency within RACER-based systems. The combination of RACER and PLATO is able to maintain a throughput of above 600 mb/s on a 100-node network, significantly outperforming the compared consensus mechanisms in terms of network node size and performance.

Distributed notification systems serve as critical infrastructure in modern digital applications, delivering time-sensitive information across environments where reliability directly impacts business operations and user experience. This article explores strategies for ensuring high availability in notification systems, addressing challenges that arise from hardware failures, network outages, and scheduled maintenance. The discussion covers foundational redundancy approaches by examining key architectural patterns, including active-active and active-passive configurations that eliminate single points of failure. The article extends to state management techniques employing consensus algorithms like Raft, Paxos, and ZAB alongside various replication strategies that balance consistency and availability requirements. Fault detection mechanisms such as heartbeat protocols, gossip protocols, and health checks are presented with graceful degradation strategies that maintain essential functionality during disruptions. Storage practices, proactive monitoring techniques, and disaster recovery planning complete the holistic approach to building resilient notification infrastructures that deliver uninterrupted service even under adverse conditions.

Opportunistic networks are characterized by intermittent connectivity and dynamic topologies, which pose significant challenges for efficient message delivery, resource management, and routing decision-making. This paper introduces the Fuzzy Control Routing Protocol, a novel approach designed to address these challenges by leveraging fuzzy logic to enhance routing decisions and improve overall network performance. The protocol considers buffer occupancy, angle to destination, and the number of unique connections of the target nodes to make context-aware routing decisions. It was implemented and evaluated using the FuzzyC framework for simulations and the opportunistic network environment simulator for realistic network scenarios. Simulation results demonstrate that the Fuzzy Control Routing Protocol achieves competitive delivery probability, efficient resource utilization, and low overhead compared to the Epidemic and MaxProp protocols. Notably, it consistently outperformed the Epidemic protocol across all metrics and exhibited comparable delivery probability to MaxProp while maintaining significantly lower overhead, particularly in low-density scenarios. The results demonstrate the protocol’s ability to adapt to varying network conditions, effectively balance forwarding and resource management, and maintain robust performance in dynamic vehicular environments.

Preventing local transmission of malaria from imported cases is crucial for achieving and maintaining malaria elimination. This study aimed to investigate the epidemiological characteristics of imported malaria cases and assess the distribution of malaria vectors in Qatar. Data from January 2016 to December 2022 on imported malaria, including demographic and epidemiological characteristics, travel-related information, and diagnostic results, were collected and analysed using descriptive statistics. Field surveys conducted in 2021-22 collected mosquitoes using various traps across Qatar. The collected samples underwent morphological and molecular characterization at Qatar University. A total of 2693 cases were reported, with a mean incidence of 13.5/100000 population, decreasing from 18.8/100000 in 2016 to 5.5/100000 in 2020. Most cases were Plasmodium vivax (57.4%) followed by P. falciparum (40.4%). The median age was 32.9 ± 12.5 years, primarily males (86.7%), expatriates (99.6%) and notified during the hot months (July to September). Cases were mainly imported from the Eastern Mediterranean Region followed by the African and South-East Asia Region with no deaths and indigenous cases. Anopheles stephensi was identified as a widely distributed species, but none carried the Plasmodium pathogen. Despite no reports of local transmission, the presence of An. stephensi and favourable environmental conditions pose a risk in Qatar. Strengthening surveillance for imported malaria and reviewing epidemic protocols are necessary. Conventional field studies are imperative to address knowledge gaps in Anopheles mosquito ecology and biting habits in Qatar, accurately assessing the risk of local malaria transmission to support Qatar's malaria-free status.

Providers of high-availability data stores need to roll out software updates without causing noticeable downtimes. For distributed data stores like Redis Cluster, the state-of-the-art is a rolling update, where the nodes are restarted in sequence. This requires preserving, restoring, and resynchronizing the database state, which can significantly prolong updates for larger memory states, and thus delay critical security fixes. In this article, we propose applying software updates directly in memory without restarting any nodes. We present the first fully operational live patching solution for Redis Cluster on Linux. We support both push- and pull-based distribution of patches, trading dissemination speed against cluster elasticity, the ability to allow nodes to dynamically join or leave the cluster. Our integration is very lightweight, as it piggybacks on the cluster-internal gossip protocol. Our experiments benchmark live patching against state-of-the-art rolling updates. In one scenario, live patching updates the entire cluster orders of magnitude faster, without unfavorable trade-offs regarding throughput, tail latencies, or network consumption. To showcase generalizability, we provide general guidelines on integrating live patching for distributed database systems and successfully apply them to a primary-replica PostgreSQL setup. Given our overall promising results, we discuss the opportunities of live patching in database DevOps.

Gossip-based asynchronous algorithms for distributed composite optimization

Due to its partially decentralized and highly scalable features, the consortium blockchain has currently overtaken other blockchain technologies as the one most frequently used and studied across various industries. However, performance issues such as low transaction efficiency and redundant communication processes continue to hinder the development of consortium blockchains. In the Hyperledger Fabric consortium blockchain system, transaction efficiency is largely influenced by the consensus protocol and broadcast protocol. This paper introduces a novel consortium blockchain network model, DC-SoC, focused on optimizing broadcast protocols. By incorporating the concept of density clustering, a stable propagation structure is established for the blockchain network, thus optimizing data dissemination in the Gossip protocol. Additionally, the concept of social networks is integrated, using trustworthiness scores and economic incentives to evaluate node security. The experimental results demonstrate that when DC-SoC is applied in a large-scale consortium blockchain environment, it significantly improves communication performance between nodes and ensures transmission reliability.

This research aims to understand the effect of the number of nodes on the performance of data communication in Nomad clusters using the gossip protocol. Through a series of tests, it can be concluded that data communication performance is greatly affected by the number of nodes in the cluster. Tests were conducted using two clusters, where one cluster consists of three nodes. The results show that when using a cluster with three nodes, no packet loss occurs in all data transmissions performed, indicating a reliable communication system. The average latency in one data communication cycle varied in each test, but generally remained within the acceptable range of below 100ms based on data communication quality of service parameters. CPU and disc usage remained relatively stable throughout the experiment. Although there were slight differences in throughput between clusters, the throughput generally remained above 100 Mbps, which is still in the good category according to the research parameters. These results show the importance of taking into account the number of nodes in the cluster in designing and managing data communication systems in a Nomad cluster environment with the gossip protocol.

Gossip-based distributed stochastic mirror descent for constrained optimization

Federated learning (FL) has emerged as a promising strategy for collaboratively training complicated machine learning models from different medical centers without the need of data sharing. However, the traditional FL relies on a central server to orchestrate the global model training among clients. This makes it vulnerable to the failure of the model server. Meanwhile, the model trained based on the global data property may not yield the best performance on the local data of a particular site due to the variations of data characteristics among them. To address these limitations, we proposed Gossip Mutual Learning(GML), a decentralized collaborative learning framework that employs Gossip Protocol for direct peer-to-peer communication and encourages each site to optimize its local model by leveraging useful information from peers through mutual learning. On the task of tumor segmentation on PET/CT images using HECKTOR21 dataset with 223 cases from five clinical sites, we demonstrated GML could improve tumor segmentation performance in terms of Dice Similarity Coefficient (DSC) by 3.2%, 4.6% and 10.4% on site-specific testing cases as compared to three baseline methods: pooled training, FedAvg and individual training, respectively. We also showed GML has comparable generalization performance as pooled training and FedAvg when applying them on 78 cases from two out-of-sample sites where no case was used for model training. In our experimental setup, GML showcased a sixfold decrease in communication overhead compared to FedAvg, requiring only 16.67% of the total communication overhead.

The peer-to-peer (P2P) networks are made up of networked devices that share resources without the need for a centralized server. The drawbacks of centralized messaging systems include their dependence on the internet, privacy issues, and censorship. Utilizing Bluetooth and Wi-Fi Direct, the system identifies nearby devices and relays messages across multiple hops in a network, employing an efficient hybrid gossip protocol (First Push, Then Pull) for optimal message propagation. The system is built with Flutter/Dart for cross-platform compatibility and features a user-friendly interface in decentralized network, including real-time updates and device management. It emphasizes scalability, fault tolerance, and privacy, making it a practical solution for communication in challenging scenarios. The implementation uses a hybrid gossip protocol, which combines the advantages of push and pull strategies, to ensure optimal message transmission. RSA (Rivest Shamir Adleman) encryption is used for end-to-end security, and offline storage ensures delivery even when recipients are temporarily unavailable. The innovative solution offers a secure and efficient alternative to conventional messaging platforms, which makes it particularly helpful in remote locations.

The e-book services we use today have a serious drawback in that we will no longer be able to read the books we have purchased when the service is terminated. One way to solve this problem is to build a decentralized system that does not depend on a specific company or organization by combining smart contracts running on the Ethereum blockchain and distributed storage such as an IPFS. However, a simple combination of existing technologies does not make the stored e-book data persistent, so the risk of purchased e-books becoming unreadable remains. In this paper, we propose a decentralized distributed storage called d-book-repository, which has both access management function and data durability for purchased e-books. This system uses NFTs as access rights to realize strict access control by preventing clients who do not have NFTs from downloading e-book data. In addition, e-book data stored on storage nodes in the distributed storage is divided into shards using Reed–Solomon codes, and each storage node stores only a single shard, thereby preventing the creation of nodes that can restore the entire content from locally stored data. The storage of each shard is not handled by a single node but by a group of nodes, and the shard is propagated to all nodes in the group using the gossip protocol, where erasure codes are utilized to increase the resilience against node departure. Furthermore, an incentive mechanism to encourage participation as a storage node is implemented using smart contracts. We built a prototype of the proposed system on AWS and evaluated its performance. The results showed that both downloading and uploading 100 MB of e-book data (equivalent to one comic book) were completed within 10 s using an instance type of m5.xlarge. This value is only 1.3 s longer for downloading and 2.2 s longer for uploading than the time required for a simple download/upload without access control, confirming that the overhead associated with the proposed method is sufficiently small.

Federated Learning (FL) enables collaborative model training among medical centers without sharing private data. However, traditional FL risks on server failures and suboptimal performance on local data due to the nature of centralized model aggregation. To address these issues, we present Gossip Mutual Learning (GML), a decentralized framework that uses Gossip Protocol for direct peer-to-peer communication. In addition, GML encourages each site to optimize its local model through mutual learning to account for data variations among different sites. For the task of tumor segmentation using 146 cases from four clinical sites in BraTS 2021 dataset, we demonstrated GML outperformed local models and achieved similar performance as FedAvg with only 25% communication overhead.

The service function (SF) area has gained increasing attention in the last years due its ability to combine the advantages of cloud computing with network softwarization. By decoupling SFs from the physical equipment where they are executed, it is possible to make network services scalable and flexible. These advantages become even more evident in the forthcoming 6G networks, where the overall environment is expected to become more dynamic and cloud-based, with SFs deployed as cloud-native functions. However, in order to efficiently manage and compose services using these SFs, it is necessary to monitor the available resources of the nodes where they can be deployed, in addition to exchange information relevant to the operational status of active SFs. To this aim, we propose a lightweight monitoring architecture by using agents in charge of monitoring the status of SFs running in co-located clusters. These monitoring agents exchange their information by means of a gossip protocol, which allows increasing the reliability of the process. In this way, it is possible to keep service decisions as local as possible, limiting the interactions with centralized decision and orchestration platforms, and thus increasing network scalability and responsiveness. Performance evaluation shows the effectiveness of the proposed solution, and demonstrates that the network overhead of the distributed monitoring process is definitely affordable.