CN-122019448-A - Method and device for calculating multi-server node time delay information through parallel processing

CN122019448ACN 122019448 ACN122019448 ACN 122019448ACN-122019448-A

Abstract

The invention relates to the technical field of software development and discloses a method and a device for parallel processing and calculating multi-server node time delay information, wherein the method comprises the steps of creating a plurality of communication examples for a target server through a transfer server node information set deployed by a back-end server, wherein the target server is a plurality of servers with communication requirements, the transfer server node information set comprises a plurality of transfer server node information, and each communication example corresponds to one transfer server node information; according to the transit server node information corresponding to all communication examples of all target servers, matching the communication examples of all target servers to obtain one or more communication example matching groups; and calculating the time delay information of the server node corresponding to each communication instance matching group by acquiring the time stamp information of each communication instance matching group. Therefore, by implementing the method and the device, the time delay information of the calculation server node can be quickened by a multi-communication-instance mode.

Inventors

ZHOU ZHENBIN
LOU KE
LI JIANFENG
BIN CHAOLIN
YANG FANFAN

Assignees

深圳鼎匠科技有限公司

Dates

Publication Date: 20260512
Application Date: 20251212

Claims (10)

1. A method for computing multi-server node delay information in parallel, the method comprising: Establishing a plurality of communication examples for a target server through a transfer server node information set deployed by a back-end server, wherein the target server is a plurality of servers with communication requirements, the transfer server node information set comprises a plurality of transfer server node information, and each communication example corresponds to one transfer server node information; According to the transit server node information corresponding to all the communication examples of all the target servers, matching all the communication examples of the target servers to obtain one or more communication example matching groups; and calculating the time delay information of the server node corresponding to each communication instance matching group by acquiring the time stamp information of each communication instance matching group.
2. The method for computing multi-server node latency information according to claim 1, wherein all the target servers include at least a first server and a second server, all the communication instances of the first server are denoted as first communication instances, and all the communication instances of the second server are denoted as second communication instances; the transfer server node information comprises node identification of a transfer server node and a key value pair of the transfer server node; And matching all the communication instances of the target server according to the transit server node information corresponding to all the communication instances of all the target servers to obtain one or more communication instance matching groups, including: Integrating all the first communication examples and all the second communication examples according to node identifiers of transit server nodes corresponding to all the first communication examples and node identifiers corresponding to all the second communication examples to obtain one or more communication example similarity groups, wherein all the communication examples in the same communication example similarity group correspond to the same node identifiers; and screening all the communication instance similarity groups meeting preset matching conditions from all the communication instance similarity groups according to key value pairs corresponding to all the communication instances in all the communication instance similarity groups to serve as communication instance matching groups.
3. The method for computing multi-server node delay information by parallel processing according to claim 2, wherein the selecting, as the communication instance matching group, all communication instance similarity groups satisfying a preset matching condition from all the communication instance similarity groups according to key value pairs corresponding to all communication instances in all the communication instance similarity groups includes: for each communication instance similarity group, judging whether a first communication instance and a second communication instance in the communication instance similarity group meet a preset matching condition according to a key value pair corresponding to the communication instance similarity group; When the first communication instance and the second communication instance in the communication instance similarity group are judged to not meet the matching condition, the communication instance similarity group is eliminated; and when the first communication instance and the second communication instance in the communication instance similarity group are judged to meet the matching condition, determining the communication instance similarity group as a communication instance matching group.
4. A method of computing multi-server node latency information in parallel processing according to claim 3, wherein the key-value pairs comprise domain name key-value pairs and/or channel key-value pairs; And for each communication instance similarity group, determining whether the first communication instance and the second communication instance in the communication instance similarity group meet a preset matching condition according to the key value pair corresponding to the communication instance similarity group, including: for each communication instance similarity group, judging whether the domain name key value pair of the first communication instance and the domain name key value pair of the second communication instance in the communication instance similarity are the same according to the domain name key value pair corresponding to the communication instance similarity group, and recording the first condition; For each communication instance similarity group, judging whether the channel key value pair of the first communication instance and the channel key value pair of the second communication instance in the communication instance similarity are the same according to the channel key value pair corresponding to the communication instance similarity group, and recording the channel key value pair as a second condition; And when at least one of the first condition and the second condition is not met, determining that the first communication instance and the second communication instance in the communication instance similarity group do not meet a preset matching condition.
5. The method for computing delay information of multiple server nodes according to any one of claims 1 to 4, wherein computing delay information of a server node corresponding to each communication instance matching group by obtaining timestamp information of each communication instance matching group comprises: determining a sending server used as a sending object and a receiving server used as a receiving object from each target server according to the communication requirements of all the target servers; establishing a communication connection between the sending server and the receiving server with respect to each communication instance matching group according to the communication instance matching group; and calculating the time delay information of the server node corresponding to each communication instance matching group by acquiring the time stamp information of the sending server and the receiving server when the communication connection of each communication instance matching group is established.
6. The method for computing delay information of multiple server nodes according to claim 5, wherein computing delay information of a server node corresponding to each communication instance matching group by acquiring timestamp information of the sending server and the receiving server when establishing a communication connection of each communication instance matching group comprises: For each communication instance matching group, acquiring a first timestamp of a heartbeat message sent by the sending server to the receiving server by the communication instance in the communication instance matching group; acquiring a second timestamp of a response message sent to the sending server by the receiving server after the heartbeat message is received by the communication instance in the communication instance matching group; And calculating the difference value between the first time stamp and the second time stamp to obtain the time delay information of the server node corresponding to the communication instance matching group.
7. The method for computing multi-server node latency information through parallel processing according to any of claims 1,2, 3,4 and 6, wherein the creating a plurality of communication instances for a target server through a set of relay server node information deployed by a backend server comprises: Deploying a transit server node information set by a back-end server, and sending the transit server node information set to a target server; Receiving the transfer server node information set sent by the back-end server by the target server, and creating a plurality of target subtasks of a target main task of the target server according to the transfer server node information set; Allocating, by the target server, at least one relay server node information from the set of relay server node information for each of the target subtasks; and creating all communication instances corresponding to each target subtask by the target server according to all transit server node information distributed for each target subtask.
8. An apparatus for computing multi-server node latency information in parallel processing, the apparatus comprising: The system comprises a creation module, a communication module and a communication module, wherein the creation module is used for creating a plurality of communication examples for a target server through a transfer server node information set deployed by a back-end server, the target server is a plurality of servers with communication requirements, the transfer server node information set comprises a plurality of transfer server node information, and each communication example corresponds to one transfer server node information; The matching module is used for matching all the communication examples of the target servers according to the transit server node information corresponding to all the communication examples of the target servers to obtain one or more communication example matching groups; and the calculation module is used for calculating the time delay information of the server node corresponding to each communication instance matching group by acquiring the time stamp information of each communication instance matching group.
9. An apparatus for computing multi-server node latency information in parallel processing, the apparatus comprising: a memory storing executable program code; a processor coupled to the memory; The processor invokes the executable program code stored in the memory to perform the method of computing multi-server node latency information in parallel processing as claimed in any one of claims 1-7.
10. A computer storage medium storing computer instructions which, when invoked, are operable to perform a method of parallel processing computing multi-server node latency information according to any of claims 1-7.

Description

Method and device for calculating multi-server node time delay information through parallel processing Technical Field The invention relates to the technical field of software development, in particular to a method and a device for parallel processing and calculating multi-server node time delay information. Background Along with the expansion of the scale of the cloud computing and the distributed system, the number of server nodes is increased rapidly, and delay information of each server node needs to be acquired, so that the response speed and the service quality of the communication system are calculated. Currently, a centralized or serial processing mode is adopted to calculate the time delay information of each server, which is a common calculation mode of the time delay information of the server node. However, the practice discovers that the existing time delay information calculation mode simply collects state data of all servers in sequence by a single management node and sequentially completes time delay calculation, or the time delay information is uniformly reported to the management node after being independently calculated according to all servers, so that the calculation speed of the time delay information is easy to be slow, the system is long in learning period and poor in instantaneity of the time delay information, and the requirement of fast sensing of the node performance under a dynamic network environment is difficult to be met. It is important to propose a technical solution for accelerating the calculation of the delay information of the server node. Disclosure of Invention The invention provides a method and a device for parallel processing of computing multi-server node time delay information, which can accelerate the computing of the time delay information of a server node, are beneficial to meeting the requirement of fast sensing of node performance in a dynamic network environment and are beneficial to improving the response speed and the service quality speed and the efficiency of a computing communication system. In order to solve the technical problem, a first aspect of the present invention discloses a method for parallel processing and calculating delay information of multiple server nodes, the method comprising: Establishing a plurality of communication examples for a target server through a transfer server node information set deployed by a back-end server, wherein the target server is a plurality of servers with communication requirements, the transfer server node information set comprises a plurality of transfer server node information, and each communication example corresponds to one transfer server node information; According to the transit server node information corresponding to all the communication examples of all the target servers, matching all the communication examples of the target servers to obtain one or more communication example matching groups; and calculating the time delay information of the server node corresponding to each communication instance matching group by acquiring the time stamp information of each communication instance matching group. As an optional implementation manner, in the first aspect of the present invention, all the target servers include at least a first server and a second server, all the communication instances of the first server are denoted as first communication instances, and all the communication instances of the second server are denoted as second communication instances; the transfer server node information comprises node identification of a transfer server node and a key value pair of the transfer server node; And matching all the communication instances of the target server according to the transit server node information corresponding to all the communication instances of all the target servers to obtain one or more communication instance matching groups, including: Integrating all the first communication examples and all the second communication examples according to node identifiers of transit server nodes corresponding to all the first communication examples and node identifiers corresponding to all the second communication examples to obtain one or more communication example similarity groups, wherein all the communication examples in the same communication example similarity group correspond to the same node identifiers; and screening all the communication instance similarity groups meeting preset matching conditions from all the communication instance similarity groups according to key value pairs corresponding to all the communication instances in all the communication instance similarity groups to serve as communication instance matching groups. In an optional implementation manner, in a first aspect of the present invention, the selecting, according to key value pairs corresponding to all communication instances in all the communication instance similarity groups, all communication instance similarity grou