Granted Patent B2

US 12,405,966 · App. 18/128,144 · Granted Sep 2, 2025

Inventor: Zenghui Bao (Shenzhen, CN)

Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED

G06F16/258G06F7/14G06F16/245

Abstract

Claims (88)

1. A data processing method performed by a computer network acting as a central cluster, the central cluster comprising multiple computer devices, the central cluster is communicatively connected to multiple edge clusters that are configured to dynamically start and destroy instances served by the central cluster, and each computer device being configured to run at least one of a transceiver component, a coordinated write component, and a database component, the transceiver component including an indicator proxy node; and the method comprising:

establishing, by the indicator proxy node, a persistent network connection to an edge cluster of the multiple edge clusters and maintaining, by the indicator proxy node, the persistent connection with the edge cluster via heartbeat packets;

receiving, by the transceiver component via the persistent network, collection indicator data sent by the edge cluster;

performing pre-aggregation processing on the collection indicator data to obtain pre-aggregated indicator data, including:

acquiring indicator information of the collection indicator data, wherein the collection indicator data includes round trip time (RTT) data and the indicator information includes an indicator attribute tag for the RTT data, and acquiring an indicator aggregation attribute;

performing indicator conversion on at least a subset of the collection indicator data that includes the round trip time (RTT) data whose indicator attribute tag matches the indicator aggregation attribute, to obtain the pre-aggregated indicator data, wherein the pre-aggregated indicator data has a size smaller than that of the subset of the collection indicator data; and

sending the pre-aggregated indicator data to the coordinated write component;

converting, by the coordinated write component, the pre-aggregated indicator data into conversion indicator data that has a target storage format supported by the database component;

performing, by the coordinated write component, merging processing on the conversion indicator data to obtain storage indicator data; and

writing, by the database component, the storage indicator data into a storage disk.

2. The method according to claim 1 , wherein the receiving, by using the transceiver component, collection indicator data sent by an edge cluster comprises:

establishing a connection to a service endpoint in the edge cluster by using the transceiver component, acquiring endpoint information of the service endpoint, and generating a connection random number based on the endpoint information;

acquiring a connection establishment range; and

receiving, when the connection random number is within the connection establishment range, collection indicator data sent by the service endpoint in the edge cluster.

3. The method according to claim 1 , wherein the performing merging processing on the conversion indicator data to obtain storage indicator data comprises:

performing downsampling processing on the conversion indicator data to obtain downsampling indicator data; and

acquiring a conversion storage period of the conversion indicator data, and performing combination processing on downsampling indicator data that has a same conversion storage period to obtain storage indicator data.

4. The method according to claim 1 , wherein the writing, by the database component, the storage indicator data into a storage disk comprises:

acquiring a storage indicator name and a storage indicator attribute tag of the storage indicator data, and generating a storage hash value of the storage indicator data based on the storage indicator name and the storage indicator attribute tag; and

determining a storage fragment identifier based on the storage hash value, and writing the storage indicator data into the database component corresponding to the storage fragment identifier.

5. The method according to claim 1 , wherein the writing, by the database component, the storage indicator data into a storage disk comprises:

performing, by the database component, cold and hot analysis processing on the storage indicator data to obtain hot indicator data and cold indicator data that are comprised in the storage indicator data; and

determining a hot data storage area and a cold data storage area in the storage disk, writing the hot indicator data into the hot data storage area, and writing the cold indicator data into the cold data storage area.

6. The method according to claim 1 , wherein the receiving, by using the transceiver component, collection indicator data sent by an edge cluster comprises:

receiving, by using the transceiver component, initial indicator data sent by the edge cluster; and

acquiring a to-be-collected indicator type, and generating collection indicator data based on indicator data corresponding to the to-be-collected indicator type and acquired from the initial indicator data.

7. The method according to claim 1 , wherein the computer device is further configured to run an indicator query component, and the method further comprises:

receiving, by the central cluster, indicator query information sent by user equipment, and sending an indicator query request to the indicator query component based on the indicator query information, the indicator query request comprising the indicator query information;

splitting, by the indicator query component, the indicator query request into M fragment query requests, M being a positive integer;

determining a query database component based on the M fragment query requests, and acquiring query indicator fragment data respectively corresponding to the M fragment query requests from the query database component; and

integrating M pieces of query indicator fragment data to obtain an indicator query result, and sending the indicator query result to the user equipment.

8. A computer network acting as a central cluster, the central cluster comprising multiple computer devices, the central cluster is communicatively connected to multiple edge clusters that are configured to dynamically start and destroy instances served by the central cluster, each computer device comprising a processor, a memory, and a computer program being configured to run at least one of a transceiver component that includes an indicator proxy node, a coordinated write component, and a database component and collectively causing the central cluster to perform a data processing method including:

establishing, by the indicator proxy node, a persistent network connection to an edge cluster of the multiple edge clusters and maintaining, by the indicator proxy node, the persistent connection with the edge cluster via heartbeat packets;

receiving, by the transceiver component via the persistent network, collection indicator data sent by the edge cluster;

performing pre-aggregation processing on the collection indicator data to obtain pre-aggregated indicator data, including:

acquiring indicator information of the collection indicator data, wherein the collection indicator data includes round trip time (RTT) data and the indicator information includes an indicator attribute tag for the RTT data, and acquiring an indicator aggregation attribute;

performing indicator conversion on at least a subset of the collection indicator data that includes the round trip time (RTT) data whose indicator attribute tag matches the indicator aggregation attribute, to obtain the pre-aggregated indicator data, wherein the pre-aggregated indicator data has a size smaller than that of the subset of the collection indicator data; and

sending the pre-aggregated indicator data to the coordinated write component;

converting, by the coordinated write component, the pre-aggregated indicator data into conversion indicator data that has a target storage format, and performing merging processing on the conversion indicator data to obtain storage indicator data; and

writing, by the database component, the storage indicator data into a storage disk.

9. The computer network according to claim 8 , wherein the receiving, by using the transceiver component, collection indicator data sent by an edge cluster comprises:

establishing a connection to a service endpoint in the edge cluster by using the transceiver component, acquiring endpoint information of the service endpoint, and generating a connection random number based on the endpoint information;

acquiring a connection establishment range; and

receiving, when the connection random number is within the connection establishment range, collection indicator data sent by the service endpoint in the edge cluster.

10. The computer network according to claim 8 , wherein the performing merging processing on the conversion indicator data to obtain storage indicator data comprises:

performing downsampling processing on the conversion indicator data to obtain downsampling indicator data; and

acquiring a conversion storage period of the conversion indicator data, and performing combination processing on downsampling indicator data that has a same conversion storage period to obtain storage indicator data.

11. The computer network according to claim 8 , wherein the writing, by the database component, the storage indicator data into a storage disk comprises:

acquiring a storage indicator name and a storage indicator attribute tag of the storage indicator data, and generating a storage hash value of the storage indicator data based on the storage indicator name and the storage indicator attribute tag; and

determining a storage fragment identifier based on the storage hash value, and writing the storage indicator data into the database component corresponding to the storage fragment identifier.

12. The computer network according to claim 8 , wherein the writing, by the database component, the storage indicator data into a storage disk comprises:

performing, by the database component, cold and hot analysis processing on the storage indicator data to obtain hot indicator data and cold indicator data that are comprised in the storage indicator data; and

determining a hot data storage area and a cold data storage area in the storage disk, writing the hot indicator data into the hot data storage area, and writing the cold indicator data into the cold data storage area.

13. The computer network according to claim 8 , wherein the receiving, by using the transceiver component, collection indicator data sent by an edge cluster comprises:

receiving, by using the transceiver component, initial indicator data sent by the edge cluster; and

acquiring a to-be-collected indicator type, and generating collection indicator data based on indicator data corresponding to the to-be-collected indicator type and acquired from the initial indicator data.

14. The computer network according to claim 8 , wherein the computer device is further configured to run an indicator query component, and the method further comprises:

receiving, by the central cluster, indicator query information sent by user equipment, and sending an indicator query request to the indicator query component based on the indicator query information, the indicator query request comprising the indicator query information;

splitting, by the indicator query component, the indicator query request into M fragment query requests, M being a positive integer;

determining a query database component based on the M fragment query requests, and acquiring query indicator fragment data respectively corresponding to the M fragment query requests from the query database component; and

integrating M pieces of query indicator fragment data to obtain an indicator query result, and sending the indicator query result to the user equipment.

15. A non-transitory computer readable storage medium, storing a plurality of computer programs being configured to run, respectively, a transceiver component that includes an indicator proxy node, a coordinated write component, and a database component, and each computer program being loaded and executed by a processor of a respective one of a plurality of computer devices constituting a computer network acting as a central cluster communicatively connected to multiple edge clusters that are configured to dynamically start and destroy instances served by the central cluster, causing the central cluster to perform a data processing method including:

establishing, by the indicator proxy node, a persistent network connection to an edge cluster of the multiple edge clusters and maintaining, by the indicator proxy node, the persistent connection with the edge cluster via heartbeat packets;

receiving, by the transceiver component via the persistent network, collection indicator data sent by the edge cluster;

performing pre-aggregation processing on the collection indicator data to obtain pre-aggregated indicator data, including:

acquiring indicator information of the collection indicator data, wherein the collection indicator data includes round trip time (RTT) data and the indicator information includes an indicator attribute tag for the RTT data, and acquiring an indicator aggregation attribute;

performing indicator conversion on at least a subset of the collection indicator data that includes the round trip time (RTT) data whose indicator attribute tag matches the indicator aggregation attribute, to obtain the pre-aggregated indicator data, wherein the pre-aggregated indicator data has a size smaller than that of the subset of the collection indicator data; and

sending the pre-aggregated indicator data to the coordinated write component;

converting, by the coordinated write component, the pre-aggregated indicator data into conversion indicator data that has a target storage format, and performing merging processing on the conversion indicator data to obtain storage indicator data; and

writing, by the database component, the storage indicator data into a storage disk.

16. The non-transitory computer readable storage medium according to claim 15 , wherein the receiving, by using the transceiver component, collection indicator data sent by an edge cluster comprises:

establishing a connection to a service endpoint in the edge cluster by using the transceiver component, acquiring endpoint information of the service endpoint, and generating a connection random number based on the endpoint information;

acquiring a connection establishment range; and

receiving, when the connection random number is within the connection establishment range, collection indicator data sent by the service endpoint in the edge cluster.

17. The non-transitory computer readable storage medium according to claim 15 , wherein the performing merging processing on the conversion indicator data to obtain storage indicator data comprises:

performing downsampling processing on the conversion indicator data to obtain downsampling indicator data; and

acquiring a conversion storage period of the conversion indicator data, and performing combination processing on downsampling indicator data that has a same conversion storage period to obtain storage indicator data.

18. The non-transitory computer readable storage medium according to claim 15 , wherein the writing, by the database component, the storage indicator data into a storage disk comprises:

performing, by the database component, cold and hot analysis processing on the storage indicator data to obtain hot indicator data and cold indicator data that are comprised in the storage indicator data; and

determining a hot data storage area and a cold data storage area in the storage disk, writing the hot indicator data into the hot data storage area, and writing the cold indicator data into the cold data storage area.

19. The non-transitory computer readable storage medium according to claim 15 , wherein the receiving, by using the transceiver component, collection indicator data sent by an edge cluster comprises:

receiving, by using the transceiver component, initial indicator data sent by the edge cluster; and

acquiring a to-be-collected indicator type, and generating collection indicator data based on indicator data corresponding to the to-be-collected indicator type and acquired from the initial indicator data.

20. The non-transitory computer readable storage medium according to claim 15 , wherein the computer device is further configured to run an indicator query component, and the method further comprises:

receiving, by the central cluster, indicator query information sent by user equipment, and sending an indicator query request to the indicator query component based on the indicator query information, the indicator query request comprising the indicator query information;

splitting, by the indicator query component, the indicator query request into M fragment query requests, M being a positive integer;

determining a query database component based on the M fragment query requests, and acquiring query indicator fragment data respectively corresponding to the M fragment query requests from the query database component; and

integrating M pieces of query indicator fragment data to obtain an indicator query result, and sending the indicator query result to the user equipment.

Assignments (1)

Priority Claims (1)

Continuity (2)

References Cited (17)