Mobility of versioned workload partitions

• US 8,464,104 B2
• Filed: 09/10/2010
• Issued: 06/11/2013
• Est. Priority Date: 09/10/2010
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Ganesh, Perinkulam I., McConaughy, John M., Ramalingam, Kavitha
• Primary Examiner(s): PATEL, KAMINI B
• Application Number:
  US12/879,812
• Publication Number:
  US 20120066555A1
• Time in Patent Office: 1,005 Days
• Field of Search: 714/49
• US Class Current: 714/49
• CPC Class Codes: G06F 11/0715 in a system implementing multitasking multitasking per se G06F9/46
• CPC Class Codes: G06F 11/0769 Readable error formats, e.g. cross-platform generic formats, human understandable formats
• CPC Class Codes: G06F 11/203 using migration
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/4856 resumption being on a different machine, e.g. task migration, virtual machine migration G06F9/5088 takes precedence
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455

Abstract

A versioned workload partition (WPAR) can be migrated from a source machine to a destination machine. Each thread associated with a process executing within the versioned WPAR is frozen. For each thread associated with the process, an error number associated with the thread is received in response to freezing execution of the thread and at least a current state of the thread is determined as checkpoint information associated with the thread based, at least in part, on the error number associated with the thread. The checkpoint information associated with the one or more threads is provided to the destination machine. The checkpoint information is used at the destination machine to reconstruct the process within a destination versioned WPAR on the destination machine.

Claims

1. A method comprising:

determining, at a source machine associated with a global operating system, that a workload partition associated with a first version of an operating system is to be migrated from the source machine to a destination machine, wherein the first version of the operating system is an earlier version than the global operating system;

halting execution of one or more threads associated with a process executing within the workload partition to be migrated from the source machine to the destination machine;

for each of the one or more threads associated with the process executing within the workload partition,

receiving an error number associated with the thread in response to said halting the execution of the thread;

determining at least a current state of the thread as checkpoint information associated with the thread based, at least in part, on the error number associated with the thread; and

providing the checkpoint information associated with each of the one or more threads associated with the process executing within the workload partition to the destination machine to migrate the workload partition from the source machine to the destination machine.

2. The method of claim 1, further comprising:

providing the error number associated with each of the one or more threads associated with the process to the destination machine.

3. The method of claim 1, wherein said determining at least the current state of the thread as checkpoint information associated with each of the one or more threads associated with the process executing within the workload partition comprises:

determining, from a kernel of the global operating system at the source machine, a memory address that references process-specific data for a process executing within the workload partition to be migrated from the source machine to the destination machine;

for each of one or more threads associated with the process executing within the workload partition,

based on the memory address that references the process-specific data of the process, determining whether the error number associated with the thread indicates that a system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system, wherein the memory address references a definition for the error number associated with the thread;

determining the current state of the thread and the error number associated with the thread as the checkpoint information associated with the thread in response to determining that the error number associated with the thread indicates that a system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system; and

determining the current state of the thread in response to determining that the error number associated with the thread does not indicate that the system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system.

4. The method of claim 3, wherein said determining, from the kernel of the global operating system at the source machine, the memory address that references the process-specific data for the process executing within the workload partition to be migrated from the source machine to the destination machine comprises:

generating, at the source machine, a request, to the kernel of the global operating system, for the memory address that references the process-specific data for the process executing within the workload partition to be migrated from the source machine to the destination machine; and

receiving, at the source machine, the memory address that references the process-specific data for the process executing within the workload partition.

5. The method of claim 4, wherein said receiving the memory address that references the process-specific data for the process executing within the workload partition further comprises:

storing, at the source machine, the memory address that references the process-specific data for the process executing within the workload partition.

6. The method of claim 5, wherein said storing the memory address that references the process-specific data for the process executing within the workload partition further comprises:

registering, at the source machine, the memory address that references the process-specific data for the process with one or more libraries associated with the global operating system.

7. The method of claim 3, wherein the process-specific data comprises an indication of one or more error numbers for the first version of the operating system with which the process executing within the workload partition is associated.

8. The method of claim 1, wherein said determining at least the current state of the thread as checkpoint information associated with each of the one or more threads associated with the process executing within the workload partition comprises:

for each of one or more threads associated with the process executing within the workload partition,

determining whether the error number associated with the thread indicates that a system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system;

determining the current state of the thread and the error number associated with the thread as the checkpoint information associated with the thread in response to determining that the error number associated with the thread indicates that a system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system; and

determining the current state of the thread in response to determining that the error number associated with the thread does not indicate that the system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system.

9. The method of claim 1, wherein said determining that the workload partition associated with the first version of the operating system is to be migrated from the source machine associated with the global operating system to the destination machine further comprises:

identifying one or more processes executing within the workload partition to be migrated from the source machine to the destination machine, wherein the one or more processes are compiled on the first version of the operating system.

10. The method of claim 1, wherein said providing the checkpoint information associated with each of the one or more threads associated with the process executing within the workload partition to the destination machine comprises:

storing the checkpoint information associated with each of the one or more threads associated with the process prior to said providing the checkpoint information associated with each of the one or more threads associated with the process executing within the workload partition to the destination machine, or

dynamically transmitting the checkpoint information associated with each of the one or more threads associated with the process executing within the workload partition to the destination machine.

11. The method of claim 1, wherein said halting the execution of the one or more threads associated with the process comprises generating a checkpoint control signal to halt the execution of the thread.

12. A method comprising:

receiving, at a destination machine associated with a global operating system, checkpoint information associated with a process executing within a first workload partition associated with a first version of an operating system of a source machine associated with a second instance of the global operating system, wherein the first version of the operating system is an earlier version than the global operating system;

for each of one or more threads, indicated by the checkpoint information, associated with the process,

determining, at the destination machine, whether an error number associated with the thread indicates that a system call generated by the thread was interrupted during execution of the system call by a kernel of the second instance of the global operating system of the source machine, wherein the checkpoint information associated with the thread comprises the error number associated with the thread;

restoring, at the destination machine, a state of the thread when execution of the thread was halted on the source machine based on the checkpoint information, and causing the system call to be reissued to a kernel of the global operating system of the destination machine, to yield a reconstructed version of the process at the destination machine, in response to determining that the error number associated with the thread indicates that the system call generated by the thread was interrupted at the source machine; and

restoring the state of the thread when execution of the thread was halted on the source machine to yield the reconstructed version of the process at the destination machine in response to determining that the error number associated with the thread indicates that the system call generated by the thread was not interrupted at the source machine.

13. The method of claim 12, further comprising causing the reconstructed version of the process to resume execution within a second workload partition associated with a second instance of the first version of the operating system, wherein the second workload partition is implemented on the destination machine.

14. The method of claim 12, wherein said determining, at the destination machine for each of the one or more threads associated with the process, whether the error number associated with the thread indicates that a system call generated by the thread was interrupted during execution of the system call by the kernel of the second instance of the global operating system of the source machine, comprises:

determining whether the checkpoint information indicates the error number associated with the thread, or

determining whether the error number associated with the thread is equal to a predetermined error number that indicates that the system call was interrupted.

15. The method of claim 12, wherein the checkpoint information associated with the process comprises checkpoint information associated with each of the one or more threads associated with the process, wherein the checkpoint information associated with each of the one or more threads comprises at least one of the state of the thread when execution of the thread was halted on the source machine and the error number associated with the thread.

16. A computer program product for mobility of versioned workload partitions, the computer program product comprising:

a non-transitory computer readable storage medium having computer usable program code embodied therewith, the computer readable program code configured to,

determine, at a source machine associated with a global operating system, that a workload partition associated with a first version of an operating system is to be migrated from the source machine to a destination machine, wherein the first version of the operating system is an earlier version than the global operating system;

halt execution of one or more threads associated with a process executing within the workload partition to be migrated from the source machine to the destination machine;

for each of the one or more threads associated with the process executing within the workload partition,

receive an error number associated with the thread in response to the computer readable program code halting the execution of the thread;

determine at least a current state of the thread as checkpoint information associated with the thread based, at least in part, on the error number associated with the thread; and

provide the checkpoint information associated with each of the one or more threads associated with the process executing within the workload partition to the destination machine to migrate the workload partition from the source machine to the destination machine.

17. The computer program product of claim 16, wherein the computer readable program code configured to determine at least the current state of the thread as checkpoint information associated with each of the one or more threads associated with the process executing within the workload partition comprises the computer readable program code configured to:

determine, from a kernel of the global operating system at the source machine, a memory address that references process-specific data for a process executing within the workload partition to be migrated from the source machine to the destination machine;

for each of one or more threads associated with the process executing within the workload partition,

based on the memory address that references the process-specific data of the process, determine whether the error number associated with the thread indicates that a system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system, wherein the memory address references a definition for the error number associated with the thread;

determine the current state of the thread and the error number associated with the thread as the checkpoint information associated with the thread in response to the computer readable program code determining that the error number associated with the thread indicates that a system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system; and

determine the current state of the thread in response to the computer readable program code determining that the error number associated with the thread does not indicate that the system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system.

18. The computer program product of claim 17, wherein the computer readable program code configured to determine the memory address that references the process-specific data for the process executing within the workload partition to be migrated from the source machine to the destination machine comprises the computer readable program code configured to:

generate a request, to the kernel of the global operating system, for the memory address that references the process-specific data for the process executing within the workload partition to be migrated from the source machine to the destination machine; and

receive the memory address that references the process-specific data for the process executing within the workload partition.

19. A machine comprising:

a processor;

a network interface coupled with the processor; and

a checkpoint unit coupled with the process and with the network interface, the checkpoint unit operable to;

determine, at the machine associated with a global operating system, that a workload partition associated with a first version of an operating system is to be migrated from the machine to a destination machine, wherein the first version of the operating system is an earlier version than the global operating system;

halt execution of one or more threads associated with a process executing within the workload partition to be migrated from the source machine to the destination machine;

for each of one or more threads associated with the process executing within the workload partition,

receive an error number associated with the thread in response to the checkpoint unit halting the execution of the thread;

determine at least a current state of the thread as checkpoint information associated with the thread based, at least in part, on the error number associated with the thread; and

provide the checkpoint information associated with each of the one or more threads associated with the process executing within the workload partition to the destination machine to migrate the workload partition from the source machine to the destination machine.

20. The machine of claim 19, wherein the checkpoint unit operable to determine at least the current state of the thread as checkpoint information associated with each of the one or more threads associated with the process executing within the workload partition based on the memory address that references the process-specific data for the process and the error number associated with the thread comprises the checkpoint unit operable to:

determine, from a kernel of the global operating system, a memory address that references process-specific data for a process executing within the workload partition to be migrated from the machine to the destination machine

for each of one or more threads associated with the process executing within the workload partition,

based on the memory address that references the process-specific data of the process, determine whether the error number associated with the thread indicates that a system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system, wherein the memory address references a definition for the error number associated with the thread;

determine the current state of the thread and the error number associated with the thread as the checkpoint information associated with the thread in response to the checkpoint unit determining that the error number associated with the thread indicates that a system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system; and

determine the current state of the thread in response to the checkpoint unit determining that the error number associated with the thread does not indicate that the system call generated by the thread was interrupted during execution of the system call by the kernel of the global operating system.