FILE SYSTEM INDEPENDENT CONTENT AWARE CACHE

• US 20110265083A1
• Filed: 04/26/2010
• Published: 10/27/2011
• Est. Priority Date: 04/26/2010
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: DAVIS, Scott Howard
• Granted Patent: US 8,312,471 B2
• Time in Patent Office: Days
• US Class Current: 718/1
• CPC Class Codes: G06F 12/084 with a shared cache
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 2212/222 Non-volatile memory
• CPC Class Codes: G06F 2212/465 Structured object, e.g. database record

Abstract

A server supporting the implementation of virtual machines includes a local memory used for caching, such as a solid state device drive. During I/O intensive processes, such as a boot storm, a “content aware” cache filter component of the hypervisor of the server first accesses a cache structure in a content cache device to determine whether data blocks have been stored in the cache structure prior to requesting the data blocks from a networked disk array via a standard I/O stack of the hypervisor. The content aware cache filter component is implemented in an I/O virtualization layer of the standard I/O stack that sits above a file system layer of the standard I/O stack, such that any file system protocol may be implemented in the file system layer.

Claims

1. A method for obtaining data for a virtual machine in a server supporting a hypervisor for running virtual machines, the method comprising:

intercepting a read request from the virtual machine intended for transmission to a virtual drive provided by the hypervisor;

confirming that the read request corresponds to a specific content type in the virtual drive;

identifying an entry in a cache index, wherein the entry comprises a virtual machine identifier corresponding to the virtual machine, a virtual drive offset value corresponding to an offset in the read request, and a reference to a data block stored in a cache maintained in a local memory in the server;

circumventing a standard I/O stack of the hypervisor to request the data block directly from the local memory by providing address information corresponding to the reference to a driver for the local memory;

receiving the data block from the driver for the local memory; and

transmitting the data block to the virtual machine in response to the read request.

2. The method of claim 1, wherein the specific content type relates to data in a boot image in the virtual drive.

3. The method of claim 2, wherein the server is networked to a disk array that stores an actual boot image comprising data blocks of the boot image in the virtual drive.

4. The method of claim 3, wherein the standard I/O stack of the hypervisor converts read requests received from virtual machines into read operations for the disk array.

5. The method of claim 1, wherein the local memory is a content cache device in the server.

6. The method of claim 1, wherein the reference is a hash value corresponding to an entry in a hash table comprising an address to the data block in the cache.

7. The method of claim 1, wherein the steps are performed by a content aware cache filter component in an I/O virtualization layer of the standard I/O stack that sits above a file system layer of the standard I/O stack.

8. The method of claim 7, wherein the I/O virtualization layer is a SCSI virtualization layer and the file system layer is a virtual machine file system layer.

9. A method for maintaining a cache structure in a server supporting a hypervisor for running virtual machines, the method comprising:

receiving a read request from a virtual machine intended for transmission to a virtual drive by the hypervisor;

transmitting the read request to a standard I/O stack of the hypervisor, wherein the standard I/O stack converts the read request into read operations for a disk array networked to the server;

receiving a data block through the standard I/O stack from the disk array in response to the read request;

computing a hash value based on the received data block;

determining whether a hash table contains an entry comprising a hash value field matching the computed hash value of the received data block and an address reference to a second data block stored in a cache maintained in a local memory in the server; and

when the hash table contains the matching entry, inserting a new entry into a cache index, wherein the new entry comprises a virtual machine identifier corresponding to the virtual machine, a virtual drive offset value corresponding to an offset in the read request, and a reference to the second data block.

10. The method of claim 9, wherein the reference is the hash value field in the identified entry in the hash table.

11. The method of claim 9, wherein the cache index is stored in the disk array, associated with the virtual drive, and loaded into the local memory of the server upon receiving the read request.

12. The method of claim 9, wherein the read request corresponds to a boot image in the virtual drive.

13. The method of claim 9, wherein the local memory is a content cache device in the server.

14. The method of claim of claim 9, further comprising the steps of:

when the hash table does not contain a matching entry, storing the received data block in a location in the cache;

inserting a new entry into the hash table, wherein the new entry comprises the hash value of the received data block and an address reference to the location in the cache; and

inserting a new entry into a cache index file, wherein the new entry comprises a virtual machine identifier corresponding to the virtual machine, a virtual drive offset value corresponding to an offset in the read request, and a reference to the data block stored in the cache.

15. The method of claim 14, wherein the reference to the data block stored in the cache is the hash value in the inserted new entry in the hash table.

16. The method of claim 9, wherein the steps are performed by a content aware cache filter component in an I/O virtualization layer of the standard I/O stack that sits above a file system layer of the standard I/O stack.

17. The method of claim 16, wherein the I/O virtualization layer is a SCSI virtualization layer and the file system layer is a virtual machine file system layer.

18. A computer-readable storage medium including instructions that, when executed by a computer processor of a server supporting a hypervisor for running virtual machines, causes the computer processor obtain data for a virtual machine by performing the steps of:

intercepting a read request from the virtual machine intended for transmission to a virtual drive provided by the hypervisor;

confirming that the read request corresponds to a specific content type in the virtual drive;

identifying an entry in a cache index file, wherein the entry comprises a virtual machine identifier corresponding to the virtual machine, a virtual drive offset value corresponding to an offset in the read request, and a reference to a data block stored in a cache maintained in a local memory in the server;

circumventing a standard I/O stack of the hypervisor to request the data block directly from the local memory by providing address information corresponding to the reference to a driver for the local memory;

receiving the data block from the driver for the local memory; and

transmitting the data block to the virtual machine in response to the read request.

19. The computer-readable storage medium of claim 18, wherein the reference to the data block is a hash value corresponding to an entry in a hash table comprising an address to the data block in the cache.

20. The computer-readable storage medium of claim 18, further including instructions that, when executed by the computer processor, performs the steps of:

receiving a second read request from a second virtual machine intended for transmission to the virtual drive;

transmitting the second read request to the standard I/O stack, wherein the standard I/O stack converts the read request into read operations for a disk array networked to the server;

receiving a second data block through the standard I/O stack from the disk array in response to the second read request;

computing a hash value based on the received second data block;

identifying an entry in a hash table wherein the entry comprises a hash value field matching the computed hash value and an address reference to a third data block stored in a cache maintained in the local memory in the server;

inserting a new entry into a cache index, wherein the new entry comprises a virtual machine identifier corresponding to the second virtual machine, a virtual drive offset value corresponding to an offset in the second read request, and a reference to the third data block stored in the cache.

21. The computer-readable storage medium of claim 20, wherein the reference to the third data block stored in the cache is the hash value field in the identified entry in the hash table.

22. The computer-readable storage medium of claim 20, further including instructions that, when executed by the computer processor, performs the steps of:

receiving a third read request from a third virtual machine intended for transmission to the virtual drive;

receiving a fourth data block through the standard I/O stack from the disk array in response to the third read request;

confirming that a hash value of the fourth data block does not correspond to any entry in the hash table;

storing the received fourth data block into a location in the cache;

inserting a new entry into the hash table, wherein the new entry comprises the hash value of the fourth data block and an address reference to the location in the cache; and

adding a second new entry into the cache index, wherein the second new entry comprises a virtual machine identifier corresponding to the third virtual machine, a virtual drive offset value corresponding to an offset in the third read request, and a reference to the fourth data block stored in the cache.

23. The computer-readable storage medium of claim 22, wherein the reference to the fourth data block stored in the cache is the hash value in the inserted new entry in the hash table.

24. The computer-readable storage medium of claim 18, wherein the local memory is a content cache device in the server.

25. The computer-readable storage medium of claim 18, wherein the specific content type relates to data in a boot image in the virtual drive.