Automatic memory management (AMM)

• US 8,051,266 B2
• Filed: 06/11/2007
• Issued: 11/01/2011
• Est. Priority Date: 06/11/2007
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: DeVal, Gary J., Hrischuk, Curtis E.
• Primary Examiner(s): PORTKA, GARY J
• Application Number:
  US11/761,026
• Publication Number:
  US 20080307183A1
• Time in Patent Office: 1,604 Days
• Field of Search: None
• US Class Current: 711/168
• CPC Class Codes: G06F 12/0269 Incremental or concurrent garbage collection, e.g. in real-time systems G06F12/0261 takes precedence

Abstract

The present invention manages the execution of multiple AMM cycles to reduce or eliminate any overlap. Specifically, the present invention provides an external supervisory process to monitor the AMM behavior of VMs on one or more nodes, and intervene when coincident AMM activity appears to be imminent. If AMM patterns suggest that two VMs are likely to perform a (e.g., a major) AMM cycle simultaneously (or with significant overlap) in the near future, the supervisory process can trigger one of the VMs to AMM immediately, or at the first ‘safe’ interval prior to the predicted AMM collision. This will have the effect of desynchronizing the AMM behavior of the VMs and maintaining AMM latency for both VMs within the expected bounds for their independent operation, without any inter-VM effects.

Claims

1. A method for Automatic Memory Management (AMM) coordination, comprising:

predicting when a first AMM cycle on a node and a second AMM cycle on the node will occur;

determining whether an overlap of the first AMM cycle with the second AMM cycle will occur;

determining whether one of the first AMM cycle and the second AMM cycle can be rescheduled to an open AMM slot if the overlap will occur; and

upon determining that the first AMM cycle and the second AMM cycle cannot be rescheduled to an open AMM slot to prevent the overlap, throttling workload to the node to defer the first AMM cycle beyond a current predicted time for the first AMM cycle.

2. The method of claim 1, wherein

the first AMM cycle being implemented by a first Virtual Machine (VM) on the node and

the second AMM cycle being implemented by second VM on the node.

3. The method of claim 1, further comprising

utilizing historical data for at least one of the predicting and the determining.

4. The method of claim 3, wherein

the historical data comprising next AMM cycle times that specify when the first AMM cycle and the second AMM cycle will commence.

5. The method of claim 4, wherein

the historical data further comprising at least one AMM duration time that specifies a duration of the first AMM cycle and a duration of the second AMM cycle, the determining further comprising using the next AMM cycle times and the at least one AMM duration time to determine whether the overlap will occur.

6. The method of claim 5, wherein

the determining whether one of the first AMM cycle and the second AMM cycle can be rescheduled being based on a AMM guard time that specifies a desired delay between termination of the first AMM cycle and commencement of the second AMM cycle.

7. The method of claim 1, wherein

the method being implemented by a system that is external to the node.

8. A system for Automatic Memory Management (AMM) coordination, comprising:

a module for predicting when a first virtual machine on a node will enter a first AMM cycle and a second virtual machine on the node will enter a second AMM cycle;

a module for determining whether an overlap of the first AMM cycle with the second AMM cycle will occur; and

a module for determining whether one of the first AMM cycle and the second AMM cycle can be rescheduled to an open AMM slot if the overlap will occur; and

a module for, upon determining that the first AMM cycle and the second AMM cycle cannot be rescheduled to an open AMM slot to prevent the overlap, throttling workload to the node to defer the first AMM cycle beyond a current predicted time for the first AMM cycle.

9. The system of claim 8, wherein

the first virtual machine and the second virtual machine both being Virtual Machines (VMs).

10. The system of claim 8, wherein

the system utilizing historical data.

11. The system of claim 10, wherein

the historical data comprising next AMM cycle times that specify when the first AMM cycle and the second AMM cycle will commence.

12. The system of claim 11, wherein

the historical data further comprising at least one AMM duration time that specifies a duration of the first AMM cycle and a duration of the second AMM cycle, the module for determining using the next AMM cycle times and the at least one AMM duration time to determine whether the overlap will occur.

13. The system of claim 12, further comprising

a module for rescheduling utilizing a AMM guard time that specifies a desired delay between the first AMM cycle and the second AMM cycle to reschedule the at least one of the first AMM cycle and the second AMM cycle,

the module for rescheduling being configured to reschedule the at least one of the first AMM cycle and the second AMM cycle if the first AMM cycle is predicted to terminate less than the AMM guard time before the second AMM cycle is predicted to commence.

14. The system of claim 8, wherein

the system being implemented external to the node.

15. A program product stored on a computer readable storage medium for Automatic Memory Management (AMM) coordination, the computer readable medium comprising program code for causing a computer system to:

predict when a first virtual machine on a node will enter a first AMM cycle and a second virtual machine on the node will enter a second AMM cycle;

determine whether an overlap of the first AMM cycle with the second AMM cycle will occur; and

determine whether one of the first AMM cycle and the second AMM cycle can be rescheduled to an open AMM slot if the overlap will occur; and

upon determining that the first AMM cycle and the second AMM cycle cannot be rescheduled to an open AMM slot to prevent the overlap, throttle workload to the node to defer the first AMM cycle beyond a current predicted time for the first AMM cycle.

16. The program product of claim 15, wherein

the first virtual machine and the second virtual machine both being Java Virtual Machines (VMs).

17. The program product of claim 15, the computer readable medium further comprising program code for causing the computer system to obtain and utilize historical data.

18. The program product of claim 15, wherein

the historical data comprising next AMM cycle times that specify when the first AMM cycle and the second AMM cycle will commence.

19. The program product of claim 18, wherein

the historical data further comprising at least one AMM duration time that specifies a duration of the first AMM cycle and a duration of the second AMM cycle, the computer readable medium further comprising program code for causing the computer system to use the next AMM cycle times and the at least one AMM duration time to determine whether the overlap will occur.

20. The program product of claim 19, the computer readable medium further comprising program code for causing the computer system to utilize a AMM guard time that specifies a desired delay between the first AMM cycle and the second AMM cycle to reschedule the at least one of the first AMM cycle and the second AMM cycle, the computer readable medium further comprising program code for causing the computer system to reschedule the at least one of the first AMM cycle and the second AMM cycle if the first AMM cycle is predicted to terminate less than the AMM guard time before the second AMM cycle is predicted to commence.

21. The program product of claim 15, wherein

the program product being implemented external to the node.

22. A method for deploying a module for Automatic Memory Management (AMM) coordination, comprising:

providing a computer infrastructure being operable to;

predict when a first virtual machine on a node will enter a first AMM cycle and a second virtual machine on the node will enter a second AMM cycle;

determine whether an overlap of the first AMM cycle with the second AMM cycle will occur; and

determine whether one of the first AMM cycle and the second AMM cycle can be rescheduled to an open AMM slot if the overlap will occur; and

upon determining that the first AMM cycle and the second AMM cycle cannot be rescheduled to an open AMM slot to prevent the overlap, throttle workload to the node to defer the first AMM cycle beyond a current predicted time for the first AMM cycle.