Method for Processing Data, Computing Node, and System

• US 20140137136A1
• Filed: 11/15/2013
• Published: 05/15/2014
• Est. Priority Date: 11/15/2012
• Status: Active Grant

• Current Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Inventors: Zhang, Tianhu
• Granted Patent: US 9,009,735 B2
• Time in Patent Office: Days
• US Class Current: 719/318
• CPC Class Codes: G06F 9/542 Event management; Broadcasting; Multicasting; Notifications

Abstract

Embodiments of the present invention provide a method for processing data, a computing node, and a system. The method includes: registering, by a BPE, an algorithm with a CEP instance; transferring, by the CEP instance when detecting that an event concerned by the algorithm satisfies a computation-triggering condition, an event required for computation to the BPE; obtaining, by the BPE, a computation result, and if determining that a further computation is required for the computation result, writing the computation result as an intermediate event to the CEP instance; and transferring, by the CEP instance when detecting that an event concerned by another algorithm satisfies a computation-triggering condition thereof and the intermediate event is an event required for computation thereof, the intermediate event to a BPE that registers the another algorithm. The CEP instance performs association of multiple events and multiple algorithms, which simplifies a computation process and improves timeliness.

Claims

1. A method for processing data, wherein the method is applied to a business processor entity (BPE), the method comprising:

registering a predefined algorithm of the BPE with a complex event processing (CEP) instance located on a computing node which is the same as the computing node which the BPE is located on, wherein the predefined algorithm of the BPE comprises an identifier of a concerned event, an identifier of an event required for computation, and a computation-triggering condition, wherein the identifier of the concerned event is used to identify an event concerned by the predefined algorithm, and wherein the identifier of the event required for computation is used to identify an event required for computation of the predefined algorithm;

obtaining the event that is required for computation of the predefined algorithm and that is transferred by the CEP instance, wherein the event required for computation of the predefined algorithm is transferred by the CEP instance when detecting that the event concerned by the predefined algorithm satisfies the computation-triggering condition;

performing a computation on the event required for computation of the predefined algorithm according to the predefined algorithm to obtain a computation result; and

when determining, according to predefined service logic, that a further computation needs to be performed on the computation result, writing the computation result as an intermediate event to the CEP instance, so that the CEP instance transfers the intermediate event to a BPE which is located on the computing node and registers the another algorithm with the CEP instance, when detecting that an event concerned by another registered algorithm other than the predefined algorithm satisfies a computation-triggering condition in the another algorithm, and the intermediate event is an event required for computation of the another algorithm.

2. The method according to claim 1, wherein after performing the computation on the event required for computation of the predefined algorithm according to the predefined algorithm to obtain the computation result, the method further comprises transferring the computation result to a data storage processor entity (DPE) so that the DPE saves the computation result to a database, wherein the DPE is located on the computing node or another computing node.

3. The method according to claim 2, wherein transferring the computation result to the data storage processor entity DPE comprises:

determining, according to the predefined service logic, whether the computation result needs to be summed; and

when the computation result needs to be summed, transferring the computation result to an arithmetic processor entity (APE) so that the APE performs a summation computation on the computation result of the BPE and a computation result of another BPE and transfers a corresponding summation result to the DPE,

wherein the APE is located on the computing node or another computing node, and

wherein the APE and the DPE are located on the same computing node or on different computing nodes.

4. A method for processing data, wherein the method is applied to a complex event processing (CEP) instance, the method comprising:

registering a predefined algorithm of a business processor entity (BPE) located on a computing node which is the same as the computing node which the CEP is located on, wherein the predefined algorithm of the BPE comprises an identifier of a concerned event, an identifier of an event required for computation, and a computation-triggering condition, wherein the identifier of the concerned event is used to identify an event concerned by the predefined algorithm, and wherein the identifier of the event required for computation is used to identify an event required for computation of the predefined algorithm;

when detecting that the event concerned by the predefined algorithm satisfies the computation-triggering condition, transferring the event required for computation of the predefined algorithm to the BPE, so that the BPE performs a computation on the event required for computation of the predefined algorithm according to the predefined algorithm to obtain a computation result;

obtaining the computation result written by the BPE, wherein the computation result is written as an intermediate event by the BPE when determining, according to predefined service logic, that a further computation needs to be performed on the computation result; and

when detecting that an event concerned by another registered algorithm other than the predefined algorithm satisfies a computation-triggering condition in the another algorithm and the intermediate event is an event required for computation of the another algorithm, transferring the intermediate event to a BPE which is located on the computing node and registers the another algorithm with the CEP instance.

5. The method according to claim 4, wherein the event concerned by the predefined algorithm comprises an original event concerned by the predefined algorithm and an intermediate event concerned by the predefined algorithm, and before detecting that the event concerned by the predefined algorithm satisfies the computation-triggering condition, the method further comprises:

extracting the original event concerned by the predefined algorithm from a message bus; and

detecting, by the CEP instance, whether the original event concerned by the predefined algorithm and the intermediate event concerned by the predefined algorithm satisfy the computation-triggering condition.

6. A method for processing data, wherein the method is applied to a computing node comprising a business processor entity (BPE) and a complex event processing (CEP) instance, the method comprising:

registering, by the BPE, a predefined algorithm of the BPE with the CEP instance, wherein the predefined algorithm comprises an identifier of a concerned event, an identifier of an event required for computation, and a computation-triggering condition, wherein the identifier of the concerned event is used to identify an event concerned by the predefined algorithm, and wherein the identifier of the event required for computation is used to identify an event required for computation of the predefined algorithm;

transferring, by the CEP instance when detecting that the event concerned by the predefined algorithm satisfies the computation-triggering condition, the event required for computation of the predefined algorithm to the BPE;

performing, by the BPE, a computation on the event required for computation of the predefined algorithm according to the predefined algorithm to obtain a computation result; and

when the BPE determines, according to predefined service logic, that a further computation needs to be performed on the computation result, writing the computation result as an intermediate event to the CEP instance; and

when detecting that an event concerned by another registered algorithm other than the predefined algorithm satisfies a computation-triggering condition in the another algorithm and the intermediate event is an event required for computation of the another algorithm, transferring, by the CEP instance, the intermediate event to a BPE which is located on the computing node and registers the another algorithm with the CEP instance.

7. The method according to claim 6, wherein after the performing, by the BPE, the computation on the event required for computation of the predefined algorithm according to the predefined algorithm to obtain the computation result, the method further comprises transferring, by the BPE, the computation result to a data storage processor entity (DPE) so that the DPE saves the computation result to a database, wherein the DPE is located on the computing node or another computing node.

8. The method according to claim 6, wherein before the detecting, by the CEP instance, that the event concerned by the predefined algorithm satisfies the computation-triggering condition, the method further comprises:

extracting an original event from a message bus; and

detecting, by the CEP instance, whether the event concerned by the predefined algorithm among the extracted original event and the written intermediate event satisfies the computation-triggering condition.

9. The method according to claim 8, wherein extracting the original event from the message bus comprises:

receiving original events concerned by algorithms of all BPEs on the computing node, wherein the original events are sent by a message server according to a subscription instruction of a scheduler;

or

extracting, by an event filter located on the computing node, from the message bus according to a subscription configuration of a scheduler, original events concerned by algorithms of all BPEs on the computing node, and transferring the original events to the CEP instance.

10. The method according to claim 7, wherein the transferring, by the BPE, the computation result to the data storage processor entity DPE comprises:

determining, by the BPE according to the predefined service logic, whether the computation result needs to be summed;

when the computation result needs to be summed, transferring the computation result to an arithmetic processor entity (APE) so that the APE performs a summation computation on the computation result of the BPE and a computation result of another BPE and transfers the summation result to the DPE,

wherein the APE is located on the computing node or another computing node, and

wherein the APE and the DPE are located on the same computing node or on different computing nodes.

11. The method according to claim 10, wherein after the performing, by the APE, the summation computation on the computation result of the BPE and a computation result of another BPE, the method further comprises:

writing, by the APE, the summation result as an intermediate event to a CEP instance located on a computing node which is the same as the computing node which the APE is located on when the APE determines, according to the predefined service logic, that a further computation needs to be performed on the summation result.

12. A computing node, comprising:

a business processor entity (BPE); and

a complex event processing (CEP) instance,

wherein the BPE is configured to register a predefined algorithm of the BPE with the CEP instance,

wherein the predefined algorithm comprises an identifier of a concerned event, an identifier of an event required for computation, and a computation-triggering condition,

wherein the identifier of the concerned event is used to identify an event concerned by the predefined algorithm,

wherein the identifier of the event required for computation is used to identify an event required for computation of the predefined algorithm,

wherein the CEP instance is configured to transfer, when detecting that the event concerned by the predefined algorithm satisfies the computation-triggering condition, the event required for computation of the predefined algorithm to the BPE,

wherein the BPE is further configured to perform a computation on the event required for computation of the predefined algorithm according to the predefined algorithm to obtain a computation result, and when determining, according to predefined service logic, that a further computation needs to be performed on the computation result, write the computation result as an intermediate event to the CEP instance; and

wherein the CEP instance is further configured to transfer the intermediate event to a BPE which is located on the computing node and registers the another algorithm with the CEP instance when detecting that an event concerned by another registered algorithm other than the predefined algorithm satisfies a computation-triggering condition in the another algorithm and the intermediate event is an event required for computation of the another algorithm.

13. The computing node according to claim 12, wherein the BPE is further configured to send the computation result to a data storage processor entity (DPE) on another computing node so that the DPE saves the computation result to a database, or wherein the computing node further comprises a DPE and the BPE is further configured to send the computation result to the DPE on the same computing node so that the DPE saves the computation result to a database.

14. The computing node according to claim 12, wherein the CEP instance is further configured to:

extract an original event from a message bus; and

detect whether the event concerned by the predefined algorithm among the extracted original event and the written intermediate event satisfies the computation-triggering condition.

15. The computing node according to claim 14, wherein the CEP instance is further configured to:

receive original events concerned by algorithms of all BPEs on the computing node, wherein the original events are sent by a message server according to a subscription instruction of a scheduler;

detect whether the event concerned by the predefined algorithm among the original events and the written intermediate event satisfies the computation-triggering condition; and

when detecting that the computation-triggering condition is satisfied, transfer the event required for computation of the predefined algorithm to the BPE;

either when detecting that an event concerned by another registered algorithm other than the predefined algorithm satisfies a computation-triggering condition in the another algorithm and the intermediate event is an event required for computation of the another algorithm, transfer the intermediate event to a BPE which is located on the computing node and registers the another algorithm with the CEP instance;

or

wherein the computing node further comprises an event filter, wherein the event filter is configured to extract original events concerned by algorithms of all BPEs on the computing node from the message bus according to a subscription configuration of the scheduler, and transfer the original events to the CEP instance; and

wherein the CEP instance is further configured to;

receive the original events transferred by the event filter;

detect whether the event concerned by the predefined algorithm among the original events and the written intermediate event satisfies the computation-triggering condition; and

when detecting that the computation-triggering condition is satisfied, transfer the event required for computation of the predefined algorithm to the BPE;

when detecting that an event concerned by another registered algorithm other than the predefined algorithm satisfies a computation-triggering condition in the another algorithm, and the intermediate event is an event required for computation of the another algorithm, transfer the intermediate event to a BPE which is located on the computing node and registers the another algorithm with the CEP instance.

16. The computing node according to claim 12, wherein either the BPE is further configured to determine, according to predefined service logic, whether the computation result needs to be summed, and when the computation result needs to be summed, send the computation result to an arithmetic processor entity (APE) on another computing node so that the APE performs a summation computation on the computation result of the BPE and a computation result of another BPE, and transfers a summation result to the DPE, or the computing node further comprises an APE, and the BPE is further configured to determine, according to predefined service logic, whether the computation result needs to be summed, and when the computation result needs to be summed, send the computation result to the APE on the same computing node so that the APE performs a summation computation on the computation result of the BPE and a computation result of another BPE and transfers a summation result to the DPE, and wherein the APE and the DPE are located on the same computing node or on different computing nodes.

17. The computing node according to claim 16, wherein the APE is further configured to write the summation result as an intermediate event to a CEP instance on the same computing node when determining, according to the predefined service logic, that a further computation needs to be performed on the summation result.

18. A distributed computing system, comprising:

a scheduler; and

at least one computing node, wherein each computing node comprises a business processor entity (BPE) and a complex event processing (CEP) instance,

wherein the BPE is configured to register a predefined algorithm of the BPE with the CEP instance,

wherein the predefined algorithm comprises an identifier of a concerned event, an identifier of an event required for computation, and a computation-triggering condition,

wherein the identifier of the concerned event is used to identify an event concerned by the predefined algorithm, and wherein the identifier of the event required for computation is used to identify an event required for computation of the predefined algorithm,

wherein the CEP instance is configured to transfer the event required for computation of the predefined algorithm to the BPE when detecting that the event concerned by the predefined algorithm satisfies the computation-triggering condition,

wherein the BPE is further configured to perform a computation on the event required for computation of the predefined algorithm according to the predefined algorithm to obtain a computation result, and when determining, according to predefined service logic, that a further computation needs to be performed on the computation result, write the computation result as an intermediate event to the CEP instance; and

wherein when detecting that an event concerned by another registered algorithm other than the predefined algorithm satisfies a computation-triggering condition in the another algorithm and the intermediate event is an event required for computation of the another algorithm, the CEP instance is further configured to transfer the intermediate event to a BPE which is located on the computing node and registers the another algorithm with the CEP instance, and

wherein the scheduler is configured to collect resource information of each computing node, determine a scheduling mode for each computing node according to the resource information, and respectively schedule each computing node according to the scheduling mode of each computing node.

19. The system according to claim 18, wherein the scheduling mode comprises a predefined algorithm, predefined service logic, and determining whether to enable a business processor entity (BPE), an arithmetic processor entity (APE), or a data storage processor entity (DPE) on the computing node.

20. The system according to claim 18, wherein there are multiple computing nodes, and wherein the scheduler is further configured to:

reserve a computing node as a standby computing node;

schedule other computing nodes than the standby computing node; and

when one of the other computing nodes that are scheduled becomes abnormal, schedule the standby computing node according to a scheduling mode of the abnormal computing node and cached data of the abnormal computing node.