Computer Device, Method, and Apparatus for Scheduling Business Flow

• US 20160260040A1
• Filed: 05/12/2016
• Published: 09/08/2016
• Est. Priority Date: 11/14/2013
• Status: Active Grant

• Current Assignee: Cac, LLC
• Original Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Inventors: Zhou, Junjie, Murthy, Vemana Narasimha, Haveri, Upasana
• Granted Patent: US 10,453,010 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 9/4843 by program, e.g. task dispatcher, supervisor, operating system
• CPC Class Codes: G06Q 10/06311 Scheduling, planning or task assignment for a person or group

Abstract

A computer device, a method, and an apparatus for business flow scheduling. The method for business flow scheduling includes determining that a transaction savepoint is configured on a to-be-scheduled node, creating an asynchronous task according to the transaction savepoint, where the asynchronous task includes a to-be-scheduled task that is on the to-be-scheduled node and that is after the transaction savepoint, using a first thread to execute a transaction that includes a task in the flow other than the asynchronous task, and submitting the transaction executed by the first thread in order to ensure data consistency, and using a second thread to execute the asynchronous task in order to implement transaction splitting between task nodes.

Claims

1. A flow scheduling apparatus configured to execute a flow, wherein the flow comprises at least one node, and wherein the flow scheduling apparatus comprises:

a processor configured to;

determine that a transaction savepoint is configured on a to-be-scheduled node;

create an asynchronous task according to the transaction savepoint, wherein the asynchronous task comprises a to-be-scheduled task configured on the to-be-scheduled node, and wherein the to-be-scheduled task is scheduled after the transaction savepoint;

execute, by a first thread, a transaction comprising a first task in the flow, wherein the first task and the asynchronous task are different;

submit the transaction executed by the first thread; and

execute, by a second thread, the asynchronous task.

2. The flow scheduling apparatus according to claim 1, wherein the processor is further configured to place the asynchronous task into a to-be-scheduled task queue of a first flow engine process, and wherein the first thread belongs to the first flow engine process.

3. The flow scheduling apparatus according to claim 2, wherein the processor is further configured to place the asynchronous task into the to-be-scheduled task queue of the first flow engine process when the asynchronous task is created by listening according to a registered transaction submission event callback listener indicating that the transaction executed by the first thread is submitted successfully.

4. The flow scheduling apparatus according to claim 2, wherein the processor is further configured to:

remove the asynchronous task from the to-be-scheduled task queue when the second thread terminates exceptionally before execution of the asynchronous task is completed; and

execute, by a third thread, the asynchronous task.

5. The flow scheduling apparatus according to claim 1, wherein the processor is further configured to:

persist the asynchronous task;

store the persisted asynchronous task in a database after the transaction executed by the first thread is submitted;

acquire, by a second flow engine process, the asynchronous task from the database when a first flow engine process terminates exceptionally before execution of the asynchronous task is completed, wherein the second thread belongs to the first flow engine process;

schedule, by the second flow engine process, the asynchronous task; and

execute, by the second flow engine process, the asynchronous task.

6. The flow scheduling apparatus according to claim 1, wherein the processor is further configured to store the transaction savepoint in a flow definition in a database.

7. The flow scheduling apparatus according to claim 1, wherein the transaction is a declarative transaction.

8. A method for scheduling a business flow, wherein the flow comprises at least one node, and wherein the method comprises:

determining that a transaction savepoint is configured on a to-be-scheduled node;

creating an asynchronous task according to the transaction savepoint, wherein the asynchronous task comprises a to-be-scheduled task configured on the to-be-scheduled node, and wherein the to-be-scheduled task is scheduled after the transaction savepoint;

executing, by a first thread, a transaction comprising a first task in the flow, wherein the first task and the asynchronous task are different;

submitting the transaction executed by the first thread; and

executing, by a second thread, the asynchronous task.

9. The method according to claim 8, further comprising placing the asynchronous task into a to-be-scheduled task queue of a first flow engine process after submitting the transaction executed by the first thread, wherein the first thread belongs to the first flow engine process.

10. The method according to claim 9, further comprising registering a transaction submission event callback listener, wherein placing the asynchronous task into the to-be-scheduled task queue of the first flow engine process further comprises placing the asynchronous task into the to-be-scheduled task queue of the first flow engine process when the asynchronous task is created by listening according to the transaction submission event callback listener indicating that the transaction executed by the first thread is successfully submitted.

11. The method according to claim 9, further comprising:

removing the asynchronous task from the to-be-scheduled task queue when the second thread terminates exceptionally before execution of the asynchronous task is completed; and

executing, by a third thread, the asynchronous task.

12. The method according to claim 8, further comprising:

persisting the asynchronous task after the creating the asynchronous task; and

storing the persisted asynchronous task in a database after submitting the transaction.

13. The method according to claim 12, further comprising:

acquiring, by a second flow engine process, the asynchronous task from the database when a first flow engine process terminates exceptionally before execution of the asynchronous task is completed, wherein the second thread belongs to the first flow engine process;

scheduling, by the second flow engine process, the asynchronous task; and

executing, by the second flow engine process, the asynchronous task.

14. The method according to claim 8, further comprising storing the transaction savepoint in a flow definition in a database.

15. A computer device for executing flow scheduling, wherein the computer device is configured to execute a flow comprising at least one node, and wherein the computer device comprises:

a memory configured to store computer executable instructions; and

a processor coupled to the memory and configured to execute the instructions to;

determine that a transaction savepoint is configured on a to-be-scheduled node;

create an asynchronous task according to the transaction savepoint, wherein the asynchronous task comprises a to-be-scheduled task configured on the to-be-scheduled node, and wherein the asynchronous task is scheduled after the transaction savepoint;

execute, by a first thread, a transaction that comprises a first task in the flow, wherein the first task and the asynchronous task are different;

submit the transaction executed by the first thread; and

execute, by a second thread, the asynchronous task.

16. The computer device according to claim 15, wherein the processor is further configured to execute the instructions to place the asynchronous task into a to-be-scheduled task queue of a first flow engine process when place the asynchronous task is created by listening according to a registered transaction submission event callback listener indicating that the transaction executed by the first thread is submitted successfully.

17. The computer device according to claim 16, wherein the processor is further configured to execute the instructions to:

remove the asynchronous task from the to-be-scheduled task queue when the second thread terminates exceptionally before execution of the asynchronous task is completed; and

execute, by a third thread, the asynchronous task.

18. The computer device according to claim 15, wherein the processor is further configured to execute the instructions to:

persist the asynchronous task;

store the persisted asynchronous task in a database after the transaction executed by the first thread is submitted;

acquire, by a second flow engine process, the asynchronous task from the database when a first flow engine process terminates exceptionally before execution of the asynchronous task is completed, wherein the second thread belongs to the first flow engine process;

schedule, by the second flow engine process, the asynchronous task; and

execute, by the second flow engine process, the asynchronous task.