COOPERATIVE MEMORY RESOURCE MANAGEMENT FOR VIRTUALIZED COMPUTING DEVICES

• US 20130054922A1
• Filed: 08/23/2011
• Published: 02/28/2013
• Est. Priority Date: 08/23/2011
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: TUCH, Harvey, NEWELL, Craig, LAPLACE, Cyprien
• Granted Patent: US 8,738,868 B2
• Time in Patent Office: Days
• US Class Current: 711/170
• CPC Class Codes: G06F 12/0284 Multiple user address space allocation, e.g. using different base addresses interprocessor communication G06F15/163
• CPC Class Codes: G06F 12/08 in hierarchically structured memory systems, e.g. virtual memory systems
• CPC Class Codes: G06F 2209/504 Resource capping
• CPC Class Codes: G06F 2212/00 Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
• CPC Class Codes: G06F 2212/152 Virtualized environment, e.g. logically partitioned system
• CPC Class Codes: G06F 9/5016 the resource being the memory
• CPC Class Codes: G06F 9/5022 Mechanisms to release resources
• 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
• CPC Class Codes: G09G 2360/08 Power processing, i.e. workload management for processors involved in display operations, such as CPUs or GPUs
• CPC Class Codes: G09G 2360/121 using a cache memory
• CPC Class Codes: G09G 5/001 Arbitration of resources in a display system, e.g. control of access to frame buffer by video controller and/or main processor
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A computing device employs a cooperative memory management technique to dynamically balance memory resources between host and guest systems running therein. According to this cooperative memory management technique, memory that is allocated to the guest system is dynamically adjusted up and down according to a fairness policy that takes into account various factors including the relative amount of readily freeable memory resources in the host and guest systems and the relative amount of memory allocated to hidden applications in the host and guest systems.

Claims

1. A method of managing memory in a computing device having host and guest systems that share memory resources, each of the host and guest systems configured to terminate processes upon experiencing memory pressure, comprising:

determining a first difference that is based on a memory availability value in the host system and a memory availability value in the guest system;

determining a second difference that is based on an amount of memory allocated to hidden applications running in the host system and an amount of memory allocated to hidden applications running in the guest system; and

allocating memory to a process running in the guest system in accordance with a target value that is a function of the first and second differences.

2. The method of claim 1, wherein the memory availability value in each of the host and guest systems is based on a number of pages in a free memory pool.

3. The method of claim 2, wherein the memory availability value in each of the host and guest systems is also based on a number of pages in a page cache that is available for on-demand memory allocation, until the number of pages in the page cache that is available for on-demand memory allocation is less than a threshold number of pages at which the corresponding host or guest system begins terminating processes.

4. The method of claim 1, wherein the amount of memory allocated to hidden applications running in each of the host and guest systems is determined as a sum of resident set sizes of the hidden applications running therein.

5. The method of claim 1, wherein the second difference is bound by an upper limit.

6. The method of claim 5, wherein the upper limit is a function of the amount of memory allocated to hidden applications running in the host and guest systems that is larger.

7. The method of claim 1, further comprising:

updating the target value on a periodic basis,

wherein the memory allocated to the process running in the guest system is adjusted in accordance with the updated target value.

8. The method of claim 1, further comprising:

detecting a memory pressure condition in the host system and, in response thereto, updating the target value,

wherein the memory allocated to the process running in the guest system is adjusted in accordance with the updated target value.

9. The method of claim 1, further comprising:

detecting a memory pressure condition in the guest system and, in response thereto, updating the target value on a periodic basis,

wherein the memory allocated to the process running in the guest system is adjusted in accordance with the updated target value.

10. The method of claim 1, wherein the process running in the guest system is a kernel driver.

11. A non-transitory computer-readable storage medium comprising instructions to be executed in a computing device having host and guest systems that share memory resources to cause the host system to carry out the steps of:

determining a first difference that is based on a memory availability value in the host system and a memory availability value in the guest system;

determining a second difference that is based on an amount of memory allocated to hidden applications running in the host system and an amount of memory allocated to hidden applications running in the guest system; and

determining a memory allocation adjustment value for the guest system as a function of the first and second differences.

12. The non-transitory computer-readable storage medium of claim 11, wherein the memory availability value in each of the host and guest systems is based on a number of pages in a free memory pool.

13. The non-transitory computer-readable storage medium of claim 12, wherein the memory availability value in each of the host and guest systems is also based on a number of pages in a page cache that is available for on-demand memory allocation, until the number of pages in the page cache that is available for on-demand memory allocation is less than a threshold number of pages at which the corresponding host or guest system begins terminating processes.

14. The non-transitory computer-readable storage medium of claim 11, wherein the amount of memory allocated to hidden applications running in each of the host and guest systems is determined as a sum of resident set sizes of the hidden applications running therein.

15. The non-transitory computer-readable storage medium of claim 11, wherein the second difference is bound by an upper limit.

16. The non-transitory computer-readable storage medium of claim 15, wherein the upper limit is a function of the amount of memory allocated to hidden applications running in the host and guest systems that is larger.

17. A computing device comprising:

a processor in which host and guest systems are executed, and memory that is shared between the host and guest systems,

wherein the host system determines a memory allocation adjustment value for the guest system as a function of a first difference that is based on a memory availability value in the host system and a memory availability value in the guest system, and a second difference that is based on an amount of memory allocated to hidden applications running in the host system and an amount of memory allocated to hidden applications running in the guest system.

18. The computing device of claim 17, wherein the memory availability value in each of the host and guest systems is based on a number of pages in a free memory pool.

19. The computing device of claim 18, wherein the memory availability value in each of the host and guest systems is also based on a number of pages in a page cache that is available for on-demand memory allocation, until the number of pages in the page cache that is available for on-demand memory allocation is less than a threshold number of pages at which the corresponding host or guest system begins terminating processes.

20. The computing device of claim 17, wherein the amount of memory allocated to hidden applications running in each of the host and guest systems is determined as a sum of resident set sizes of the hidden applications running therein.

21. The computing device of claim 17, wherein the second difference is bound by an upper limit.

22. The computing device of claim 21, wherein the upper limit is a function of the amount of memory allocated to hidden applications running in the host and guest systems that is larger.