Terminating an accelerator application program in a hybrid computing environment

• US 9,417,905 B2
• Filed: 02/03/2010
• Issued: 08/16/2016
• Est. Priority Date: 02/03/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Archer, Charles J., Bellows, Gregory H., Burdick, Dean J., Carey, James E., Ceason, Jeffrey M., Markland, Matthew W., Sanders, Philip J., Stewart, Gordon G.
• Primary Examiner(s): Mudrick, Timothy A
• Application Number:
  US12/699,162
• Publication Number:
  US 20110191785A1
• Time in Patent Office: 2,386 Days
• Field of Search: 719/318
• US Class Current: 1/1
• CPC Class Codes: G06F 9/46 Multiprogramming arrangements

Abstract

Terminating an accelerator application program in a hybrid computing environment that includes a host computer having a host computer architecture and an accelerator having an accelerator architecture, where the host computer and the accelerator are adapted to one another for data communications by a system level message passing module (‘SLMPM’), and terminating an accelerator application program in a hybrid computing environment includes receiving, by the SLMPM from a host application executing on the host computer, a request to terminate an accelerator application program executing on the accelerator; terminating, by the SLMPM, execution of the accelerator application program; returning, by the SLMPM to the host application, a signal indicating that execution of the accelerator application program was terminated; and performing, by the SLMPM, a cleanup of the execution environment associated with the terminated accelerator application program.

Claims

1. A method of terminating an accelerator application program in a hybrid computing environment, the hybrid computing environment comprising a host computer having a host computer architecture and an accelerator having an accelerator architecture, the accelerator architecture optimized, with respect to the host computer architecture, for speed of execution of a particular class of computing functions, the host computer and the accelerator adapted to one another for data communications by a system level message passing module (‘

SLMPM’

), wherein the accelerator is configured to execute one or more accelerator application programs, the method comprising;

receiving, by the SLMPM from a host application executing on the host computer, a request to terminate an accelerator application program executing on the accelerator, wherein the request to terminate the accelerator application comprises an accelerator identifier uniquely identifying the accelerator, and a process identifier uniquely identifying the accelerator application;

terminating, by the SLMPM, execution of the accelerator application program;

returning, by the SLMPM to the host application, a signal indicating that execution of the accelerator application program was terminated; and

performing, by the SLMPM, a cleanup of the execution environment associated with the terminated accelerator application program by causing undelivered messages associated with the terminated accelerator application program to indicate an error.

2. The method of claim 1 wherein terminating, by the SLMPM, execution of the accelerator application program further comprises terminating execution of the accelerator application program without terminating the host application and without terminating other accelerator application programs executing on the accelerator.

3. The method of claim 1 wherein receiving a request to terminate an accelerator application program executing on the accelerator further comprises receiving the request to terminate the accelerator application program through an Application Programming Interface (‘

API’

) exposed to the host application by the SLMPM.

4. The method of claim 1 wherein performing a cleanup of the execution environment associated with the terminated accelerator application program further comprises performing a device cleanup including deregistering a process identifier of the terminated accelerator application program with the accelerator.

5. The method of claim 1 wherein performing a cleanup of the execution environment associated with the terminated accelerator application program further comprises performing a connection cleanup including removing mappings of a process identifier of the terminated accelerator application program to process identifiers of one or more other accelerator application programs.

6. The method of claim 1 wherein performing a cleanup of the execution environment associated with the terminated accelerator application program further comprises administering message queues associated with the terminated accelerator application program, administering execution environment objects shared by the terminated accelerator application program, and invalidating a process identifier of the terminated accelerator application program for access through the SLMPM.

7. The method of claim 1 further comprising sending, asynchronously by the host application to the SLMPM, the request to terminate the accelerator application program executing on the accelerator responsive to determining that the accelerator application program is nonresponsive.

8. The method of claim 1 wherein the host computer and the accelerator are adapted to one another for data communications by two or more data communications fabrics of at least two different fabric types.

9. A hybrid computing environment for terminating an accelerator application program in a hybrid computing environment, the hybrid computing environment comprising a host computer having a host computer architecture, an accelerator having an accelerator architecture, the accelerator architecture optimized, with respect to the host computer architecture, for speed of execution of a particular class of computing functions, the host computer and the accelerator adapted to one another for data communications by a system level message passing module, the accelerator configured to execute one or more accelerator application programs, the system level message passing module comprising computer program instructions that when executed cause a processor to carry out the steps of:

receiving, by the SLMPM from a host application executing on the host computer, a request to terminate an accelerator application program executing on the accelerator, wherein the request to terminate the accelerator application comprises an accelerator identifier uniquely identifying the accelerator, and a process identifier uniquely identifying the accelerator application;

terminating, by the SLMPM, execution of the accelerator application program;

returning, by the SLMPM to the host application, a signal indicating that execution of the accelerator application program was terminated; and

performing, by the SLMPM, a cleanup of the execution environment associated with the terminated accelerator application program by causing undelivered messages associated with the terminated accelerator application program to indicate an error.

10. The hybrid computing environment of claim 9 wherein terminating, by the SLMPM, execution of the accelerator application program further comprises terminating execution of the accelerator application program without terminating the host application and without terminating other accelerator application programs executing on the accelerator.

11. The hybrid computing environment of claim 9 wherein receiving a request to terminate an accelerator application program executing on the accelerator further comprises receiving the request to terminate the accelerator application program through an Application Programming Interface (‘

API’

) exposed to the host application by the SLMPM.

12. The hybrid computing environment of claim 9 wherein performing a cleanup of the execution environment associated with the terminated accelerator application program further comprises performing a device cleanup including deregistering a process identifier of the terminated accelerator application program with the accelerator.

13. The hybrid computing environment of claim 9 wherein performing a cleanup of the execution environment associated with the terminated accelerator application program further comprises performing a connection cleanup including removing mappings of a process identifier of the terminated accelerator application program to process identifiers of one or more other accelerator application programs.

14. The hybrid computing environment of claim 9 wherein performing a cleanup of the execution environment associated with the terminated accelerator application program further comprises administering message queues associated with the terminated accelerator application program, administering execution environment objects shared by the terminated accelerator application program, and invalidating a process identifier of the terminated accelerator application program for access through the SLMPM.

15. The hybrid computing environment of claim 9 wherein the host application further comprises computer program instructions that when executed cause a processor to carry out the steps of sending, asynchronously to the SLMPM, the request to terminate the accelerator application program executing on the accelerator responsive to determining that the accelerator application program is nonresponsive.

16. The hybrid computing environment of claim 9 wherein the host computer and the accelerator are adapted to one another for data communications by two or more data communications fabrics of at least two different fabric types.

17. A computer program product for terminating an accelerator application program in a hybrid computing environment, the hybrid computing environment comprising a host computer having a host computer architecture;

an accelerator having an accelerator architecture, the accelerator architecture optimized, with respect to the host computer architecture, for speed of execution of a particular class of computing functions, the host computer and the accelerator adapted to one another for data communications by a system level message passing module, the accelerator configured to execute one or more accelerator application programs, the computer program product disposed upon a computer readable recordable storage medium, the computer program product comprising computer program instructions that when executed cause a computer to carry out the steps of;

receiving, by the SLMPM from a host application executing on the host computer, a request to terminate an accelerator application program executing on the accelerator, wherein the request to terminate the accelerator application comprises an accelerator identifier uniquely identifying the accelerator, and a process identifier uniquely identifying the accelerator application;

terminating, by the SLMPM, execution of the accelerator application program;

returning, by the SLMPM to the host application, a signal indicating that execution of the accelerator application program was terminated; and

performing, by the SLMPM, a cleanup of the execution environment associated with the terminated accelerator application program by causing undelivered messages associated with the terminated accelerator application program to indicate an error.