Shortcut input/output in virtual machine systems

• US 9,032,181 B2
• Filed: 05/19/2010
• Issued: 05/12/2015
• Est. Priority Date: 05/19/2009
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Ahmad, Irfan, Waldspurger, Carl A.
• Primary Examiner(s): Parikh, Kalpit
• Application Number:
  US12/783,205
• Publication Number:
  US 20100299667A1
• Time in Patent Office: 1,819 Days
• Field of Search: 711/203
• US Class Current: 711/203
• CPC Class Codes: G06F 12/1018 involving hashing techniques, e.g. inverted page tables
• CPC Class Codes: G06F 12/1036 for multiple virtual address spaces, e.g. segmentation G06F12/1045 takes precedence
• CPC Class Codes: G06F 12/109 for multiple virtual address spaces, e.g. segmentation G06F12/1036 takes precedence
• CPC Class Codes: G06F 16/2255 Hash tables
• CPC Class Codes: G06F 16/273 Asynchronous replication or reconciliation
• CPC Class Codes: G06F 16/30 of unstructured textual data document management systems G06F16/93
• CPC Class Codes: G06F 16/325 Hash tables
• CPC Class Codes: G06F 2212/152 Virtualized environment, e.g. logically partitioned system
• CPC Class Codes: G06F 2212/656 Address space sharing
• CPC Class Codes: G06F 9/46 Multiprogramming arrangements
• CPC Class Codes: G06F 9/4856 resumption being on a different machine, e.g. task migration, virtual machine migration G06F9/5088 takes precedence
• 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

Read requests to a commonly accessed storage volume are conditionally issued, depending on whether or not a requested data block is already stored in memory from a prior access or to be stored in memory upon completion of a pending request. A data structure is maintained in memory to track physical memory pages and to indicate for each physical memory page the corresponding location in the storage volume from which the contents of the physical memory were read and the number of virtual memory pages that are mapped thereto.

Claims

1. A method of processing read I/O requests in a computer system having applications running therein, the applications including virtual memory pages mapped to machine memory pages, the method comprising:

maintaining a tracking data structure for a first group of machine memory pages, each having at least two virtual memory pages mapped thereto, and a second group of machine memory pages, each having only one virtual memory page mapped thereto, the tracking data structure indicating, for each of said machine memory pages, a corresponding location in a storage volume from which its contents were read or are being read;

receiving a read request;

identifying a first location of the storage volume indicated in the read request;

calculating, using the tracking data structure, a likelihood that the first location of the storage volume will be shared;

determining that the likelihood exceeds a threshold value;

based on the determination that the likelihood exceeds the threshold value;

determining, using the tracking data structure, that a machine memory page in the first group or the second group contains or will contain data stored in the first location of the storage volume indicated in the read request;

based on the determination that a machine memory page in the first group or the second group contains or will contain data stored in a location of the storage volume indicated in the read request, determining not to process the read request; and

based on the determination to not process the read request, mapping a virtual memory page associated with the read request to the machine memory page; and

in response to a first memory write request, copying the contents of a machine memory page in the first group to a first machine memory page, and mapping a virtual memory page associated with the first memory write request to the first machine memory page; and

in response to a second memory write request, copying the contents of a machine memory page in the second group to a second machine memory page, and mapping a virtual memory page associated with the second memory write request to the second machine memory page.

2. The method of claim 1, wherein the applications are instances of virtual machines and the virtual memory pages are guest physical memory pages.

3. The method of claim 1, wherein the tracking data structure indicates, for each of said machine memory pages, a reference count indicating how many virtual memory pages are mapped to the machine memory page, and, in response to said read request, the reference count associated with the machine memory page in the first group or the second group is incremented.

4. The method of claim 3, wherein, in response to said first memory write request, the reference count associated with the machine memory page in the first group is decremented, and, in response to said second memory write request, the reference count associated with the machine memory page in the second group is decremented.

5. The method of claim 4, further comprising:

updating the tracking data structure by deleting one or more of entries of the tracking data structure that correspond to the machine memory pages that have no virtual memory pages mapped thereto.

6. The method of claim 1, wherein at least one of the machine memory pages in the first and second groups is a target memory location of a read request issued to the storage volume and completed.

7. The method of claim 1, wherein at least one of the machine memory pages in the first and second groups is a target memory location of a read request issued to the storage volume and currently in process.

8. The method of claim 1, wherein calculating a likelihood that the first location of the storage volume will be shared is based on a location of the first location of the storage volume.

9. The method of claim 1, wherein calculating a likelihood that the first location of the storage volume will be shared is based on a number of times the first location of storage volume has been requested.

10. A method of processing read I/O requests in a computer system having applications running therein, the applications including virtual memory pages mapped to machine memory pages, the method comprising:

maintaining a tracking data structure for a first group of machine memory pages, each having at least two virtual memory pages mapped thereto, a second group of machine memory pages, each having only one virtual memory page mapped thereto, and a third group of machine memory pages, each having no virtual memory pages mapped thereto, the tracking data structure indicating, for each of said machine memory pages, a corresponding location in a storage volume from which its contents were read or are being read, and a reference count indicating how many virtual memory pages are mapped to the machine memory page;

receiving a read request;

identifying a first location of the storage volume indicated in the read request;

calculating, using the tracking data structure, a likelihood that the first location of the storage volume will be shared;

determining that the likelihood exceeds a threshold value; and

based on the determination that the likelihood exceeds the threshold value, mapping a virtual memory page associated with the read request to a machine memory page in one of the first, second, and third groups, and incrementing the reference count associated with the machine memory page.

11. The method of claim 10, wherein at least one of the machine memory pages in the first, second, and third groups is a target memory location of a read request issued to the storage volume and completed.

12. The method of claim 10, wherein at least one of the machine memory pages in the first, second, and third groups is a target memory location of a read request issued to the storage volume and currently in process.

13. The method of claim 10, further comprising:

updating the tracking data structure to delete one or more entries associated with the machine memory pages in the third group.

14. The method of claim 10, wherein, in response to a memory write request, copying the contents of a machine memory page in one of the first and second groups to a new machine memory page, and mapping a virtual memory page associated with the memory write request to the new machine memory page.

15. A method of processing read I/O requests in a computer system having virtual machines running therein, the virtual machines including guest virtual memory pages mapped to machine memory pages via guest physical memory pages, the method comprising:

maintaining a tracking data structure for a set of machine memory pages, the tracking data structure indicating, for each of said machine memory pages, a corresponding location in a storage volume from which its contents were read or are being read, a reference count indicating how many virtual memory pages are mapped to the machine memory page, and a pending status flag indicating whether or not a read from the storage volume is pending;

receiving a read request;

identifying a first location of the storage volume indicated in the read request;

calculating, using the tracking data structure, a likelihood that the first location of the storage volume will be shared;

determining that the likelihood exceeds a threshold value; and

based on the determination that the likelihood exceeds the threshold value;

determining, using the tracking data structure, if one of the machine memory pages being tracked contains or will contain data stored in a location of the storage volume indicated in the read request;

based on the determination one of the machine memory pages being tracked contains or will contain data stored in a location of the storage volume indicated in the read request, determining not to process the read request; and

based on the determination to not process the read request, mapping a guest physical memory page associated with the read request to the machine memory page; and

if there is no machine memory page being tracked that contains or will contain data stored in a location of the storage volume indicated in the read request, issuing a request to the storage volume for the data stored in the first location of the storage volume indicated in the read request, mapping a guest physical memory page associated with the read request to the machine memory page in which the requested data will be stored, and adding a new entry to a data structure containing page sharing hints, the new entry identifying the guest physical memory page associated with the read request and the virtual machine associated with the guest physical memory page.

16. The method of claim 15, further comprising:

adding a new entry to the tracking data structure corresponding to the machine memory page in which the requested data will be stored, the new entry including the location of the storage volume indicated in the read request, a reference count indicating that one guest physical memory pages is mapped to the machine memory page, and a pending status flag that is set to indicate whether a read from the storage volume is pending or has completed.

17. A computer system comprising:

a host platform for virtual machines; and

a storage volume connected thereto, the host platform for virtual machines including one or more processors and machine memory, the machine memory having stored therein a tracking data structure for a first group of machine memory pages, each having at least two virtual memory pages mapped thereto, a second group of machine memory pages, each having only one virtual memory page mapped thereto, and a third group of machine memory pages, each having no virtual memory pages mapped thereto, the tracking data structure indicating, for each of said machine memory pages, a corresponding location in a storage volume from which its contents were read or are being read, and a reference count indicating how many virtual memory pages are mapped to the machine memory page, and a pending status flag indicating whether or not a read from the storage volume is pending, wherein a read request issued by any of the virtual machines is conditionally issued to the storage volume based on the tracking data structure; and

wherein the one or more processors are programmed to;

receiving a read request;

identifying a first location of the storage volume indicated in the read request;

calculating, using the tracking data structure, a likelihood that the first location of the storage volume will be shared;

determining that the likelihood exceeds a threshold value; and

based on the determination that the likelihood exceeds the threshold value, map a virtual memory page associated with the read request to a machine memory page in one of the first, second, and third groups.

18. The computer system of claim 17, wherein the processors are programmed to process read requests issued by the virtual machines and, prior to issuing a read request, to check the tracking data structure to determine whether there exists a machine memory page that contains or will contain data from a location in the storage volume that is the same as indicated in the read request.

19. The computer system of claim 17, wherein the processors are programmed to map a virtual memory page to a machine memory page and increment the reference count associated with the machine memory page if it is determined that the machine memory page contains or will contain data from a location in the storage volume that is the same as indicated in the read request.

20. The computer system of claim 19, wherein the processors are further programmed to copy the contents of a shared machine memory page to a new machine memory page in response to a request to write to the shared machine memory page and decrement the reference count associated with the shared machine memory page.