Distributed scheduling in a virtual machine environment

• US 10,318,349 B2
• Filed: 05/22/2017
• Issued: 06/11/2019
• Est. Priority Date: 05/22/2017
• Status: Active Grant

• Current Assignee: Salesforce.com, Inc.
• Original Assignee: Salesforce.com, Inc.
• Inventors: Sura, Bhinav, Devaraj, Dilip, Sarkapally, Rajavardhan, Gowdru, Kirankumar Kakanuru
• Primary Examiner(s): Sun, Charlie
• Application Number:
  US15/601,199
• Publication Number:
  US 20180336054A1
• Time in Patent Office: 750 Days
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: G06F 9/4887 involving deadlines, e.g. rate based, periodic

Abstract

A first scheduler stores into a memory of a first virtual machine, a first block of jobs to be executed by the first virtual machine, the first block of jobs included in a table stored in a database associated with a server computer system. A second scheduler stores into a memory of a second virtual machine, a second block of jobs to be executed by the second virtual machine. The second block of jobs being included in the table and having a second block size equal to the first block size and including jobs not in the first block. From the first virtual machine memory, the first scheduler schedules one or more jobs in the first block for execution by the first virtual machine. From the second virtual machine memory, the second scheduler schedules one or more jobs in the second block for execution by the second virtual machine.

Claims

1. A method comprising:

storing by a first scheduler into a memory of a first virtual machine running on a server computer system, a first block of jobs to be executed by the first virtual machine, the first block of jobs being included in a job table stored in a database and having a first block size, the database associated with the server computer system;

storing by a second scheduler into a memory of a second virtual machine running on the server computer system, a second block of jobs to be executed by the by the second virtual machine, the second block of jobs being included in the job table and having a second block size, the second block size being equal to the first block size and including jobs that are not in included in the first block of jobs;

scheduling by the first scheduler from the memory of the first virtual machine, one or more jobs in the first block of jobs for execution by the first virtual machine; and

scheduling by the second scheduler from the memory of the second virtual machine, one or more jobs in the second block of jobs for execution by the second virtual machine.

2. The method of claim 1, wherein the first block of jobs is determined based on an index pointing to a job in the job table, and the first block size used to determine jobs to be included in the first block of jobs.

3. The method of claim 2, wherein the index points to a first job in the first block of jobs, and the index is updated to point to a first job in the second block of jobs after the storing of the first block of jobs into the memory of the first virtual machine.

4. The method of claim 3, further comprising comparing the index to a total number of jobs in the job table to determine whether there are jobs remaining in the job table to enable a third scheduler to store a third block of jobs included in the job table into a memory of a third virtual machine.

5. The method of claim 4, further comprising at least accessing the index and locking access to the index in order to compare the index to the total number of jobs in the job table.

6. The method of claim 5, wherein there are no remaining jobs in the job table for the third scheduler based on the index being equal to the total number of jobs in the job table.

7. The method of claim 6, wherein the scheduling of jobs by the first scheduler and the second scheduler is stopped based on an end of scheduling time.

8. The method of claim 7, wherein the storing by the first scheduler and the second scheduler is refreshed based on the end of scheduling time, the index is reset, and the end of scheduling time is updated to a next end of scheduling time based on a configurable refresh cycle.

9. The method of claim 8, wherein the second scheduler is configured to schedule the jobs in the first block of jobs based on a failover of the first virtual machine.

10. An apparatus comprising:

one or more processors; and

a non-transitory computer readable medium storing a plurality of instructions, which when executed, cause the one or more processors to;

store by a first scheduler into a memory of a first virtual machine running on a server computer system, a first block of jobs to be executed by the first virtual machine, the first block of jobs being included in a job table stored in a database and having a first block size, the database associated with the server computer system;

store by a second scheduler into a memory of a second virtual machine running on the server computer system, a second block of jobs to be executed by the second virtual machine, the second block of jobs being included in the job table and having a second block size, the second block size being equal to the first block size and including jobs that are not in included in the first block of jobs;

schedule by the first scheduler from the memory of the first virtual machine, one or more jobs in the first block of jobs for execution by the first virtual machine; and

schedule by the second scheduler from the memory of the second virtual machine, one or more jobs in the second block of jobs for execution by the second virtual machine.

11. The apparatus of claim 10, wherein the first block of jobs is determined based on an index pointing to a job in the job table and the first block size used to determine jobs to be included in the first block of jobs.

12. The apparatus of claim 11, wherein the index points to a first job in the first block of jobs, and the index is updated to point to a first job in the second block of jobs after the storing of the first block of jobs into the memory of the first virtual machine.

13. The apparatus of claim 12, further comprising instructions to compare the index to a total number of jobs in the job table to determine whether there are jobs remaining in the job table to enable a third scheduler to store a third block of jobs included in the job table into a memory of a third virtual machine.

14. The apparatus of claim 13, further comprising instructions to at least access the index and to lock access to the index in order to compare the index to the total number of jobs in the job table.

15. The apparatus of claim 14, wherein there are no remaining jobs in the job table for the third scheduler based on the index being equal to the total number of jobs in the job table.

16. The apparatus of claim 15, wherein the instructions to schedule jobs by the first scheduler and the second scheduler include instructions to stop scheduling based on an end of scheduling time.

17. The apparatus of claim 16, wherein the storing by the first scheduler and the second scheduler is refreshed based on the end of scheduling time, the index is reset, and the end of scheduling time is updated to a next end of scheduling time based on a configurable refresh cycle.

18. The apparatus of claim 17, wherein the second scheduler is configured to schedule the jobs in the first block of jobs based on a failover of the first virtual machine.

19. A computer program product comprising computer-readable program code to be executed by one or more processors when retrieved from a non-transitory computer-readable medium, the program code including instructions to:

store by a first scheduler into a memory of a first virtual machine running on a server computer system, a first block of jobs to be executed by the first virtual machine, the first block of jobs being included in a job table stored in a database and having a first block size, the database associated with the server computer system;

store by a second scheduler into a memory of a second virtual machine running on the server computer system, a second block of jobs to be executed by the second virtual machine, the second block of jobs being included in the job table and having a second block size, the second block size being equal to the first block size and including jobs that are not in included in the first block of jobs;

schedule by the first scheduler from the memory of the first virtual machine, one or more jobs in the first block of jobs for execution by the first virtual machine; and

schedule by the second scheduler from the memory of the second virtual machine, one or more jobs in the second block of jobs for execution by the second virtual machine.

20. The computer program product of claim 19, wherein the first block of jobs is determined based on an index pointing to a job in the job table.

21. The computer program product of claim 20, wherein the index points to a first job in the first block of jobs, and the index is updated to point to a first job in the second block of jobs after the storing of the first block of jobs into the memory of the first virtual machine.

22. The computer program product of claim 21, further comprising instructions to compare the index to a total number of jobs in the job table to determine whether there are jobs remaining in the job table to enable a third scheduler to store a third block of jobs included in the job table into a memory of a third virtual machine.

23. The computer program product of claim 22, further comprising instructions to at least access the index and to lock access to the index in order to compare the index to the total number of jobs in the job table.

24. The computer program product of claim 23, wherein there are no remaining jobs in the job table for the third scheduler based on the index being equal to the total number of jobs in the job table.

25. The computer program product of claim 24, wherein the instructions to schedule jobs by the first scheduler and the second scheduler include instructions to stop scheduling based on an end of scheduling time.

26. The computer program product of claim 25, wherein the storing by the first scheduler and the second scheduler is refreshed based on the end of scheduling time, the index is reset, and the end of scheduling time is updated to a next end of scheduling time based on a configurable refresh cycle.

27. The computer program product of claim 26, wherein the second scheduler is configured to schedule the jobs in the first block of jobs based on a failover of the first virtual machine.