Search Images Maps Play YouTube News Gmail Drive More »
Advanced Patent Search | Web History | Sign in

Patents

A method of reallocating memory to a hypervisor in a virtualized computing system, includes: assigning priorities to a plurality of logical partitions configured within the virtualized computing system; determining a memory requirement for the hypervisor, the hypervisor configured to manage the plurality of logical partitions; determining minimum levels of memory required for each of the plurality of logical partitions; determining the amount of available memory in the virtualized computing system; and in the event that the amount of available memory is less than the determined memory requirement of the hypervisor, removing portions of memory from one or more of the plurality of logical partitions based on the assigned priorities until the determined memory requirement for the hypervisor is obtained.

InventorsBryan M. Logan, Andres Gonzalez, Jr., Mark G. Manges
Original AssigneeInternational Business Machines Corporation
Primary Examiner: Hiep T Nguyen
Attorney: Cantor Colburn LLP
Current U.S. Classification711/173

View patent at USPTO
Search USPTO Assignment Database
Download USPTO Public PAIR data

Citations

Cited PatentFiling dateIssue dateOriginal AssigneeTitle
US20050262505Apr 25, 2005Method and apparatus for dynamic memory resource management
US20080209153Feb 28, 2007Page oriented memory management
US20080235457Mar 21, 2007Dynamic quality of service (QoS) for a shared cache

Claims

1. A method for reallocating memory to a hypervisor in a virtualized computing system, the method comprising:

assigning priorities to a plurality of logical partitions configured within the virtualized computing system;

determining a memory requirement for the hypervisor, the hypervisor configured to manage the plurality of logical partitions;

determining minimum levels of memory required for each of the plurality of logical partitions;

determining the amount of available memory in the virtualized computing system; and
in the event that the amount of available memory is less than the determined memory requirement of the hypervisor, removing portions of memory from one or more of the plurality of logical partitions based on the assigned priorities until the determined memory requirement for the hypervisor is obtained.

2. The method of claim 1, wherein memory is taken from each of the plurality of logical partitions until a minimum operating level of memory for a logical partition is reached.

3. The method of claim 1, wherein:

the plurality of logical partitions are assigned tiers, comprising a minimum memory amount, a current memory amount, and a maximum memory amount; and

the tiers are used to rank the plurality of logical partitions.

4. The method of claim 3, wherein one or more of the lowest ranked tiered partitions are depleted of memory below the minimum memory amount thereof whenever the total amount of memory obtained from the plurality of logical partitions is insufficient to meet the determined memory requirement of the hypervisor; and

successive tiers from the plurality of logical partitions are additionally depleted of memory below the minimum amount thereof until the determined memory amount of hypervisor is met.

5. The method of claim 1, wherein coefficients are assigned to each of the plurality of logical partitions;

wherein each of the assigned coefficients is proportional to the relative use of the hypervisor by each of the plurality of logical partitions.

6. A system for reallocating memory to a hypervisor in a virtualized computing system, the system comprising:

a hypervisor with a configuration manager, memory, and a plurality of logical partitions;

wherein the configuration manager assigns priorities to the plurality of logical partitions, determines a memory requirement for the hypervisor, determines minimum levels of memory required for each of the plurality of logical partitions, and determines the amount of available memory in the virtualized computing system not committed to the plurality of logical partitions; and

in the event that the amount of available memory is less than the determined memory requirement of the hypervisor, the configuration manager removes portions of memory from one or more of the plurality of logical partitions based on the assigned priorities until the determined memory requirement for the hypervisor is obtained.

7. The system of claim 6, wherein memory is taken from each of the plurality of logical partitions until a minimum operating level of memory for a logical partition is reached.

8. The system of claim 6, wherein:

the plurality of logical partitions are assigned tiers, comprising a minimum memory amount, a current memory amount, and a maximum memory amount; and

the tiers are used to rank the plurality of logical partitions.

9. The system of claim 8, wherein a lowest ranked tiered partition is depleted of memory below the minimum memory amount thereof whenever the total amount of memory obtained from the plurality of logical partitions is insufficient to meet the determined memory requirement of the hypervisor; and

successive tiers from the plurality of logical partitions are additionally depleted of memory below the minimum amount thereof until the determined memory amount of hypervisor is met.

10. The system of claim 6, wherein coefficients are assigned to each of the plurality of logical partitions;

wherein each of the assigned coefficients is proportional to the relative use of the hypervisor by each of the plurality of logical partitions.

11. An article comprising one or more computer-readable storage media containing instructions that when executed enable a computer to implement a method of reallocating memory to a hypervisor in a virtualized computing system; wherein the method further comprises:

assigning priorities to a plurality of logical partitions configured within the virtualized computing system;

determining a memory requirement for the hypervisor, the hypervisor configured to manage the plurality of logical partitions;

determining minimum levels of memory required for each of the plurality of logical partitions;

determining the amount of available memory in the virtualized computing system; and
in the event that the amount of available memory is less than the determined memory requirement of the hypervisor, removing portions of memory from one or more of the plurality of logical partitions based on the assigned priorities until the determined memory requirement for the hypervisor is obtained.

12. The article of claim 11, wherein memory is taken from each of the plurality of logical partitions until a minimum operating level of memory for a partition is reached.

13. The article of claim 11, wherein:

the plurality of logical partitions are assigned tiers, comprising a minimum memory amount, a current memory amount, and a maximum memory amount; and

the tiers are used to rank the plurality of logical partitions.

14. The article of claim 13, wherein a lowest ranked tiered partition is depleted of memory below the minimum memory amount thereof whenever the total amount of memory obtained from the plurality of logical partitions is insufficient to meet the determined memory requirement of the hypervisor; and

successive tiers from the plurality of logical partitions are additionally depleted of memory below the minimum amount thereof until the determined memory amount of hypervisor is met.

15. The article of claim 11, wherein coefficients are assigned to each of the plurality of logical partitions;

wherein each of the assigned coefficients is proportional to the relative use of the hypervisor by each of the plurality of logical partitions.