5 results about "Client server mode" patented technology

The invention discloses an asynchronous check point cache control method and device for a high-performance computing system and a medium, and the method comprises the steps: dividing computing nodes in the system into an SBB group and a PFS group, and enabling the SBB group and the PFS group to complete check point writing at the minimum time difference; a server is started at each SBB node, a client is embedded in each process of each computing node of the SBB group, and each client communicates with one SBB server; in the check point caching stage, check points of the SBB group and the PFS group are correspondingly written into the SBB and the PFS in different modes, the SBB group uses a client-server mode, and the PFS group uses a file I / O mode; in the checkpoint refreshing stage, the checkpoints in the SBB are refreshed in a client-server mode. The performance bottleneck in the large-scale application check point process in the HPC system can be effectively relieved, the check point efficiency is improved, and the influence on the calculation task is reduced.

This application discloses an asynchronous checkpoint caching control method, device, and medium for high-performance computing systems. The method includes the following steps: dividing the computing nodes in the system into SBB groups and PFS groups, and ensuring that the SBB and PFS groups complete checkpoint writing with minimal time difference; starting a server on each SBB node, embedding a client in each process of each computing node in the SBB group, and having each client communicate with an SBB server; during the checkpoint caching phase, using different modes to write checkpoints from the SBB and PFS groups to the corresponding SBB and PFS groups, respectively, where the SBB group uses a client-server mode and the PFS group uses a file I / O mode; during the checkpoint refresh phase, checkpoints in the SBB are refreshed using the client-server mode. This application can effectively alleviate performance bottlenecks during large-scale checkpoint applications in HPC systems, improve checkpoint efficiency, and reduce the impact on computing tasks.

The present invention discloses a method for collecting and labeling API encrypted traffic based on a middleman proxy, which belongs to the technical field of network traffic data collection. The present invention adopts a middleman proxy client to detect all network access requests of applications of application terminals, parses the traffic of interest, extracts the URL links, parameters and other information therein to form an API interface document; utilizes the mapping between the target application process, the target application process number and the API traffic source port to achieve the matching of the encrypted network flow with the specific application, thereby achieving the purpose of labeling the API encrypted traffic and generating the corresponding log document. The present invention improves the efficiency and scalability of traffic collection and labeling based on the client-server model, and can not only realize a distributed structure, but also collect large-scale application API encrypted traffic and corresponding keys. The present invention can automatically complete the simulation of multiple user API request parameters, and directionally generate and collect API request traffic.

The present application relates to the technical field of maritime traffic service, in particular to a kind of offshore general navigation monitoring method based on mobile and beidou short message hybrid communication, comprising the following steps: S1, communication model is built, including infrastructure and communication mode;Infrastructure is based on client / server mode;Communication mode includes mobile communication, beidou short message and hybrid communication;S2, client and server are based on strategy mode, according to communication demand and communication condition, dynamically adjust communication mode;The communication strategy includes mobile communication priority strategy, transmission strategy based on statistics and transmission strategy based on machine learning;S3, build communication process, including the division of communication stage, the setting and conversion of communication state, and the relationship between communication process.The present application supports the real-time switching of communication strategy, has the sending, receiving and management capability of monitoring data in the complex situation such as frequent disconnection, long-time network reconnection.

The application relates to a label noise robust federated learning method based on self-step learning and an adjacency matrix, a horizontal federated learning framework based on a client-server mode, and distributed training, and comprises the following steps: step S1, selecting a trusted client; step S2, calculating the near neighbor relationship of all client samples according to the trusted client; performing federated learning according to the trusted client obtained in step S1 to obtain a global federated model, and using the global federated model to calculate the near neighbor relationship of all client samples; step S3, self-step updating of sample near neighbor relationship and label evaluation and correction; and step S4, out-of-cluster sample processing. The application provides a safe and reliable label noise robust federated learning method based on self-step learning and an adjacency matrix, can effectively reduce the interference of noise data on a model, and improves the model convergence speed.