Handling frequently accessed pages

Patent metadata

Field	Value
Publication number	US-11803306-B2
Application number	US-202117236952-A
Country	US
Kind code	B2
Filing date	Apr 21, 2021
Priority date	Jun 27, 2017
Publication date	Oct 31, 2023
Grant date	Oct 31, 2023

Abstract

Handling frequently accessed pages is disclosed. An indication is received of a stalling event caused by a requested portion of memory being inaccessible. It is determined that the requested portion of memory is a frequently updated portion of memory. The stalling event is handled based at least in part on the determination that the requested portion of memory is a frequently updated portion of memory.

First claim

What is claimed is:

1. A computer system comprising:
a plurality of physically interconnected computing nodes;
a guest operating system to run in a virtual environment that is defined by a set of hyper-kernels running on the plurality of physically interconnected computing nodes; and
a hyper-kernel on a first computing node of the plurality of physically interconnected computing nodes to:
receive an indication of a stalling event caused by a requested portion of a memory being inaccessible, wherein the requested portion of the memory is requested by a virtual processor in the first computing node, and the requested portion of the memory is on a second computing node of the plurality of physically interconnected computing nodes;
determine, by the hyper-kernel performing a query of a data structure comprising identifiers of portions of the memory designated as being frequently updated, whether the requested portion of the memory is designated as a frequently updated portion of the memory; and
based at least in part on a determination that the requested portion of the memory is designated as a frequently updated portion of the memory, handle the stalling event at least in part by migrating the virtual processor to the second computing node to update the requested portion of the memory, wherein the virtual processor is to be migrated back to the first computing node in response to a completion of an update to the requested portion of the memory.

2. The computer system of claim 1 , wherein the query is performed using an address of the requested portion of the memory.

3. The computer system of claim 2 , wherein the address comprises a guest physical address.

4. The computer system of claim 1 , wherein an identifier of a portion of the memory is included in the data structure based at least in part on a frequency at which the portion of the memory is accessed.

5. The computer system of claim 4 , wherein the frequency at which the portion of the memory is accessed is determined based at least in part on a number of write stalls that have occurred with respect to the portion of the memory.

6. The computer system of claim 4 , wherein the identifier of the portion of the memory is included in the data structure based at least in part on a determination that the frequency at which the portion of the memory is accessed exceeds a threshold.

7. The computer system of claim 1 , wherein the stalling event comprises a write stall involving the virtual processor attempting to write into the requested portion of the memory.

8. The computer system of claim 1 , wherein at least some of the computing nodes of the plurality of physically interconnected computing nodes are to receive copies of the updated requested portion of the memory.

9. A method comprising:
receiving, by a hyper-kernel on a first computing node, an indication of a stalling event caused by a requested portion of a memory being inaccessible, wherein the first computing node is included in a plurality of physically interconnected computing nodes, wherein a guest operating system is run in a virtual environment that is defined by a set of hyper-kernels running on the plurality of physically interconnected computing nodes, and wherein the requested portion of the memory is requested by a virtual processor in the first computing node, and the requested portion of the memory is on a second computing node of the plurality of physically interconnected computing nodes;
determining, at least in part by the hyper-kernel performing a query of a data structure comprising identifiers of portions of the memory designated as being frequently updated, whether the requested portion of the memory is designated as a frequently updated portion of the memory;
based at least in part on a determination of that the requested portion of the memory is designated as a frequently updated portion of the memory, handling the stalling event at least in part by migrating the requested portion of the memory to the second computing node to be updated; and
migrating the requested portion of the memory back to the first computing node in response to a completion of an update to the requested portion of the memory.

10. The method of claim 9 , wherein the query is performed using an address of the requested portion of the memory.

11. The method of claim 10 , wherein the address comprises a guest physical address.

12. The method of claim 9 , wherein an identifier of a portion of the memory is included in the data structure based at least in part on a frequency at which the portion of the memory is accessed.

13. The method of claim 12 , wherein the frequency at which the portion of the memory is accessed is determined based at least in part on a number of write stalls that have occurred with respect to the portion of the memory.

14. The method of claim 12 , wherein the identifier of the portion of the memory is included in the data structure based at least in part on a determination that the frequency with which the portion of the memory is accessed exceeds a threshold.

15. The method of claim 9 , wherein the stalling event comprises a write stall involving the virtual processor attempting to write into the requested portion of the memory.

16. The method of claim 9 , comprising broadcasting copies of the updated requested portion of the memory to at least some of the computing nodes of the plurality of physically interconnected computing nodes.

17. A non-transitory computer readable storage medium comprising computer instructions that upon execution cause a computer system to:
receive, at a hyper-kernel on a first computing node, an indication of a stalling event caused by a requested portion of a memory being inaccessible, wherein the first computing node is included in a plurality of physically interconnected computing nodes, wherein a guest operating system is run in a virtual environment that is defined by a set of hyper-kernels running on the plurality of physically interconnected computing nodes, and wherein the requested portion of the memory is requested by a virtual processor in the first computing node, and the requested portion of the memory is on a second computing node of the plurality of physically interconnected computing nodes;
determine, at least in part by the hyper-kernel performing a query of a data structure comprising identifiers of portions of the memory designated as being frequently updated, whether the requested portion of the memory is designated as a frequently updated portion of the memory;
based at least in part on a determination that the requested portion of the memory is designated as a frequently updated portion of the memory, handle the stalling event at least in part by migrating the virtual processor to the second computing node to update the requested portion of the memory; and
migrate the requested portion of the memory back to the first computing node in response to a completion of an update to the requested portion of the memory.

18. The non-transitory computer readable storage medium of claim 17 , wherein an identifier of a portion of the memory is included in the data structure based at least in part on a frequency at which the portion of the memory is accessed.

19. The non-transitory computer readable storage medium of claim 18 , wherein the frequency at which the portion of the memory is accessed is determined based at least in part on a number of write stalls that have occurred with respect to the portion of the memory.

20. The non-transitory computer readable sto

Assignees

• Hewlett Packard Entpr Dev Lp

Inventors

• Nassi Isaac R
• Ioannidou Kleoni
• Berman Michael
• Fang I-Chun
• Hill Mark
• Moffet Brian
• Radick Jeffrey Paul
• Reed David P
• Reynolds Keith

Classifications

• G06F3/0611 Primary

  in relation to response time · CPC title

• G06F3/067

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title

• G06F3/0647

  Migration mechanisms · CPC title

• G06F3/0653

  Monitoring storage devices or systems · CPC title

• G06F9/4856 Primary

  resumption being on a different machine, e.g. task migration, virtual machine migration (G06F9/5088 takes precedence) · CPC title