Search Results (2)

Decentralized networks bring us many benefits, but as networks evolve, many nodes either actively or passively become unreachable behind an NAT or a firewall. This has become a hindrance to the development of decentralized networks, where peer-to-peer communication data transfer between unreachable nodes cannot be accomplished, whether in decentralized file systems, decentralized social, or decentralized IoT. The existing scheme requires a series of centralized servers or requires network-wide flooding for consensus data, which can lead to the loss of decentralized nature of the network and cause flooding bottlenecks, contrary to the design concept of decentralization. In this paper, our proposed scheme uses a structured P2P overlay network to store the indexes of unreachable nodes in the whole network, so that the characteristics of a decentralized network are still maintained while ensuring the efficiency of lookup. When nodes communicate, the transmission channel is established so that both nodes continuously transmit data streams peer-to-peer without relying on the central server. Moreover, the scheme guarantees the security and privacy of nodes’ data transmission and the P2P overlay network without relying on centralized trusted institutions. Finally, we deploy a real cluster environment to verify the effectiveness of each module at different network sizes and prove the overall feasibility of the scheme. The scheme has certain advantages over existing solutions in terms of security, privacy, communication efficiency, device democracy, etc. Full article

Distributed clustering algorithms have proven to be effective in dramatically reducing execution time. However, distributed environments are characterized by a high rate of failure. Nodes can easily become unreachable. Furthermore, it is not guaranteed that messages are delivered to their destination. As a result, fault tolerance mechanisms are of paramount importance to achieve resiliency and guarantee continuous progress. In this paper, a fault-tolerant distributed k-means algorithm is proposed on a grid of commodity machines. Machines in such an environment are connected in a peer-to-peer fashion and managed by a gossip protocol with the actor model used as the concurrency model. The fact that no synchronization is needed makes it a good fit for parallel processing. Using the passive replication technique for the leader node and the active replication technique for the workers, the system exhibited robustness against failures. The results showed that the distributed k-means algorithm with no fault-tolerant mechanisms achieved up to a 34% improvement over the Hadoop-based k-means algorithm, while the robust one achieved up to a 12% improvement. The experiments also showed that the overhead, using such techniques, was negligible. Moreover, the results indicated that losing up to 10% of the messages had no real impact on the overall performance. Full article