METHOD AND APPARATUS FOR VIRTUAL PROCESSOR DISPATCHING TO A PARTITION BASED ON SHARED MEMORY PAGES

• US 20090204959A1
• Filed: 02/07/2008
• Published: 08/13/2009
• Est. Priority Date: 02/07/2008
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Heyrman, Peter J., Anand, Vaijayanthimala K., Olszewski, Bret R.
• Granted Patent: US 8,271,989 B2
• Time in Patent Office: Days
• US Class Current: 718/1
• CPC Class Codes: G06F 2009/45583 Memory management, e.g. access or allocation
• CPC Class Codes: G06F 2212/656 Address space sharing
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• 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

The present invention provides a computer implemented method, data processing system, and computer program product for mapping and dispatching virtual processors in a data processing system having at least a first partition and a second partition. The data processing system runs a first partition on a virtual processor during a first timeslice. The data processing system identifies an at least one physical page used by the first partition and the second partition. The data processing system maps the at least one physical page to the first partition and the second partition. The data processing system determines a fitness value based on the mapping. The data processing system dispatches the Virtual processor to the second partition on a second timeslice based on the fitness value, wherein the second timeslice immediately succeeds after the first timeslice, whereby the at least one physical page remains in cache during at least the first timeslice and the second timeslice.

Claims

1. A computer implemented method for dispatching virtual processors in a data processing system having at least a first partition and a second partition, the method comprising:

running the first partition on a virtual processor during a first timeslice;

identifying an at least one physical page used by the first partition and the second partition;

mapping the at least one physical page to the first partition and the second partition;

determining a fitness value based on the mapping; and

dispatching the virtual processor to the second partition on a second timeslice based on the fitness value, wherein the second timeslice immediately succeeds after the first timeslice, whereby the at least one physical page remains in cache during at least the first timeslice and the second timeslice.

2. The computer implemented method of claim 1, wherein the fitness value is proportional to a number of physical pages mapped to the first partition and the second partition.

3. The computer implemented method of claim 2, wherein the mapping step further comprises:

matching a first physical page and a second physical page based on a hash value assigned to the first physical page, the hash value assigned to the second physical page, and a bytewise match of the first physical page and the second physical page, wherein the first physical page and the second physical page are among the at least one physical page; and

mapping a first virtual address corresponding to the first physical page to form a virtual address mapped to the second physical page, wherein at least one second virtual address maps to the second physical page.

4. The computer implemented method of claim 3, wherein the fitness value is a cache reused affinity value that relates a number of shared pages between the first virtual processor and the second virtual processor.

5. The computer implemented method of claim 4, wherein the virtual processor and the second virtual processor are part of the data processing system, and the virtual processor does not share a cache with the second virtual processor.

6. The computer implemented method of claim 1, the virtual processor shares a cache with a second virtual processor.

7. The computer implemented method of claim 1, wherein the fitness value is a pseudorandom number.

8. A computer program product comprising:

a computer usable medium having computer usable program code for dispatching virtual processors in a data processing system having at least a first partition and a second partition, said computer program product including;

computer usable program code for running a first partition on a virtual processor during a first timeslice;

computer usable program code for identifying an at least one physical page used by the first partition and the second partition;

computer usable program code for mapping the at least one physical page to the first partition and the second partition;

computer usable program code for determining a fitness value based on the mapping; and

computer usable program code for dispatching the virtual processor to the second partition on a second timeslice based on the fitness value, wherein the second timeslice immediately succeeds after the first timeslice, whereby the at least one physical page remains in cache during at least the first timeslice and the second timeslice.

9. The computer program product of claim 8, wherein the fitness value is proportional to a number of physical pages mapped to the first partition and the second partition.

10. The computer program product of claim 9, wherein the computer usable program code for mapping step further comprises:

computer usable program code for matching a first physical page and a second physical page based on a hash value assigned to the first physical page, the hash value assigned to the second physical page, and a bytewise match of the first physical page and the second physical page, wherein the first physical page and the second physical page are among the at least one physical page; and

computer usable program code for mapping a first virtual address corresponding to the first physical page to form a virtual address mapped to the second physical page, wherein at least one second virtual address maps to the second physical page.

11. The computer program product of claim 10, wherein the virtual processor does not share a cache with a second virtual processor and the fitness value is a cache reused affinity value that relates a number of shared pages between the first virtual processor and the second virtual processor.

12. The computer program product of claim 8, wherein the virtual processor and the second virtual processor are part of the data processing system, and the virtual processor does not share a cache with the second virtual processor

13. The computer program product of claim 8, wherein the fitness value is a pseudorandom number.

14. A data processing system comprising:

a bus;

a storage device connected to the bus, wherein computer usable code is located in the storage device;

a communication unit connected to the bus;

a processing unit connected to the bus, wherein the processing unit executes the computer usable code for dispatching virtual processors in the data processing system having at least a first partition and a second partition, wherein the processing unit executes the computer usable code to run the first partition on a virtual processor during a first timeslice;

identify an at least one physical page used by the first partition and the second partition;

map the at least one physical page to the first partition and the second partition;

determine a fitness value based on the mapping; and

dispatch the virtual processor to the second partition on a second timeslice based on the fitness value, wherein the second timeslice immediately succeeds after the first timeslice, whereby the at least one physical page remains in cache during at least the first timeslice and the second timeslice.

15. The data processing system of claim 14, wherein the fitness value is proportional to a number of physical pages mapped to the first partition and the second partition.

16. The data processing system of claim 15, wherein in executing computer usable code to map, the processing unit executes computer usable code to match a first physical page and a second physical page based on a hash value assigned to the first physical page, the hash value assigned to the second physical page, and a bytewise match of the first physical page and the second physical page, wherein the first physical page and the second physical page are among the at least one physical page;

and map a first virtual address corresponding to the first physical page to form a virtual address mapped to the second physical page, wherein at least one second virtual address maps to the second physical page.

17. The data processing system of claim 16, wherein the fitness value is a cache reused affinity value that relates a number of shared pages between the first virtual processor and the second virtual processor.

18. The data processing system of claim 17, wherein the virtual processor and the second virtual processor are part of the data processing system, and the virtual processor does not share a cache with the second virtual processor.

19. The data processing system of claim 14, wherein the virtual processor shares a cache with a second virtual processor.

20. The data processing system of claim 14, wherein the fitness value is a pseudorandom number.