Node performance detection system and method for decentralized storage network

[0038] A node performance detection system of a decentralized storage network, the system includes: several peer-to-peer network nodes; the network nodes are connected through a point-to-point network; the nodes include: a node detection unit, a random data access unit, a detection Response unit, data statistics recording unit and control unit; the node detection unit is used for periodic active detection of reachability and availability between nodes in the point-to-point decentralized storage network; the random data access unit is used for Perform storage access performance detection between nodes in a point-to-point decentralized storage network; the detection response unit is used to receive detection messages and respond in time; the data statistics recording unit: used for data exchange and data process, and record the process to form a statistical report of node performance; the control unit is used to construct a node ID, encrypt and sign the detection data, and the const

[0055] On the basis of the previous embodiment, the node further includes a node connection unit for:

[0056] making the node periodically obtain node ID information of other nodes;

[0057] making the node determine the degree of ID coincidence with other nodes, and retain the ID information of the relevant nodes whose ID coincidence degree is greater than a preset threshold;

[0058] determining the proximity of the node to each of the related nodes according to the ID information of the related nodes;

[0059] The nodes are made to sequentially select the related nodes with the largest proximity to connect until the maximum connection number of the nodes is reached.

[0060] Specifically, when the node ID is determined, point-to-point connection node selection: how each node selects other nodes to be connected, including the following steps:

[0061] a) Each node can set the size of the distributed hash table to store the state of some nodes in the network;

[0062] b) The

[0069] On the basis of the previous embodiment, the node detection unit is specifically used for:

[0070] Randomly select a preset number of nodes to be detected from the nodes in the distributed hash table inside the node;

[0071] From the nodes to be detected, the node to be detected with the highest detectability is sequentially selected for detection.

[0072] Specifically, point-to-point detection node selection: how each node selects the node to be detected in each round, including the following steps:

[0073] a) Selection of detection nodes From the nodes in the distributed hash table inside the node, note that this does not refer to the connected nodes;

[0074] b) Each round is selected separately and a random factor is added. This random factor can consider the time factor to ensure the randomness and detectability of each round of detection:

[0075]

[0076] select V detect The larger one is detected, and the number of detections in each round is configurable