System and method for supporting fault tolerant job management in a cloud platform environment

1. A system for supporting fault tolerant job management in a cloud computing environment, comprising:

a computer, including a cloud environment executing thereon;

one or more service management engines;

wherein each service management engine corresponds to a service type and supports provisioning of an enterprise application as a service instance of the service type on one or more clusters of the cloud environment by generating administrative commands;

a job manager that manages the execution of jobs within the cloud environment including their job states;

a job manager service that provides an application program interface which receives the administrative commands to be processed within the cloud environment as jobs;

wherein the job manager is configured to register and assign identifications to the jobs and is further configured to save or purge results of execution of the administrative commands in accordance with a retention policy; and

wherein the job manager supports fault tolerant job processing including

associating the jobs with checkpoints,

recognizing a failover command for the jobs, and

associating the jobs with work units of the administrative commands, and

storing a state for each job upon processing each work unit of the command.

2. The system of claim 1 , wherein upon failover of one or more pending jobs, the job manager automatically attempts to execute all of those pending jobs with a stored checkpoint that had been abnormally terminated.

3. The system of claim 1 , wherein each of the checkpoints indicate job attribute and status of the job at the time of creating the checkpoint, and are deleted if and when the job is completed.

4. The system of claim 1 , wherein the failover command allows particular jobs to be reverted.

5. The system of claim 1 , wherein each job unit executes and store checkpoints after each unit execution, and wherein if a failed job is reverted, job execution continues from the checkpoint associated with the last failed work unit.

6. A method for supporting fault tolerant job management in a cloud computing environment, comprising:

providing, at a computer, a cloud environment;

providing one or more service management engines,

wherein each service management engine corresponds to a service type and supports provisioning of an enterprise application as a service instance of the service type on one or more clusters of the cloud environment by generating administrative commands;

providing a job manager that manages the execution of jobs within the cloud environment including their job states;

providing a job manager service that provides an application program interface which receives the administrative commands to be processed within the cloud environment as jobs;

wherein the job manager is configured to register and assign identifications to the jobs and is further configured to save or purge results of execution of the administrative commands in accordance with a retention policy; and

wherein the job manager supports fault tolerant job processing including

associating the jobs with checkpoints,

recognizing a failover command for the jobs, and

associating the jobs with work units of the administrative commands, and storing a state for each job upon processing each work unit of the command.

7. The method of claim 6 , wherein upon failover of one or more pending jobs, the job manager automatically attempts to execute all of those pending jobs with a stored checkpoint that had been abnormally terminated.

8. The method of claim 6 , wherein each of the checkpoints indicate job attribute and status of the job at the time of creating the checkpoint, and are deleted if and when the job is completed.

9. The method of claim 6 , wherein the failover command allows particular jobs to be reverted.

10. The method of claim 6 , wherein each job unit executes and store checkpoints after each unit execution, and wherein if a failed job is reverted, job execution continues from the checkpoint associated with the last failed work unit.

11. A non-transitory computer readable storage medium, including instructions stored thereon which when read and executed by one or more computers cause the one or more computers to perform the steps comprising:

providing, at a computer, a cloud environment;

providing one or more service management engines,

wherein each service management engine corresponds to a service type and supports provisioning of an enterprise application as a service instance of the service type on one or more clusters of the cloud environment by generating administrative commands;

providing a job manager that manages the execution of jobs within the cloud environment including their job states;

providing a job manager service that provides an application program interface which receives the administrative commands to be processed within the cloud environment as jobs;

wherein the job manager is configured to register and assign identifications to the jobs and is further configured to save or purge results of execution of the administrative commands in accordance with a retention policy; and

wherein the job manager supports fault tolerant job processing including

associating the jobs with checkpoints,

recognizing a failover command for the jobs, and

associating the jobs with work units of the administrative commands, and storing a state for each job upon processing each work unit of the command.

12. The non-transitory computer readable storage medium of claim 11 , wherein upon failover of one or more pending jobs, the job manager automatically attempts to execute all of those pending jobs with a stored checkpoint that had been abnormally terminated.

13. The non-transitory computer readable storage medium of claim 11 , wherein each of the checkpoints indicate job attribute and status of the job at the time of creating the checkpoint, and are deleted if and when the job is completed.

14. The non-transitory computer readable storage medium of claim 11 , wherein the failover command allows particular jobs to be reverted.

15. The non-transitory computer readable storage medium of claim 11 , wherein each job unit executes and store checkpoints after each unit execution, and wherein if a failed job is reverted, job execution continues from the checkpoint associated with the last failed work unit.