Deduplication of virtual machine files in a virtualized desktop environment

• US 8,996,800 B2
• Filed: 10/07/2011
• Issued: 03/31/2015
• Est. Priority Date: 07/07/2011
• Status: Active Grant

• Current Assignee: Atlantis Computing, Inc. (hiveIO, Inc.)
• Original Assignee: Atlantis Computing, Inc. (hiveIO, Inc.)
• Inventors: Venkatesh, Chetan, Iyer, Kartikeya, Gaonkar, Shravan, Dixit, Sagar Shyam, Dorairajan, Vinodh
• Primary Examiner(s): LANE, JOHN A
• Application Number:
  US13/269,525
• Publication Number:
  US 20130013865A1
• Time in Patent Office: 1,271 Days
• Field of Search: 711/113, 711/135, 711/162
• US Class Current: 711/113
• CPC Class Codes: G06F 12/0808 with cache invalidating means G06F12/0815 takes precedence
• CPC Class Codes: G06F 12/126 with special data handling, e.g. priority of data or instructions, handling errors or pinning
• CPC Class Codes: G06F 12/128 adapted to multidimensional cache systems, e.g. set-associative, multicache, multiset or multilevel
• CPC Class Codes: G06F 16/172 Caching, prefetching or hoarding of files
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors

Abstract

Techniques for deduplication of virtual machine files in a virtualized desktop environment are described, including receiving data into a page cache, the data being received from a virtual machine and indicating a write operation, and deduplicating the data in the page cache prior to committing the data to storage, the data being deduplicated in-band and in substantially real-time.

Claims

1. A method, comprising:

receiving, at a deduplication application in a read-write path of a virtualized computing environment between a plurality of a virtual machines and a primary storage of the virtualized computing environment, a write operation and data associated with the write operation from a virtual machine, and a read operation and an indication of data requested with the read operation from a second virtual machine, wherein the write operation and data are received by the deduplication application in the read-write path prior to storage of the data in the primary storage of the virtualized computing environment;

receiving data into a cache memory of the deduplication application, the data being received from the deduplication application and indicating the write operation of the virtual machine, wherein the cache memory of the deduplication application is deployed in the read-write path of a virtualized computing environment; and

deduplicating the data in the cache memory by the deduplication application in substantially real-time in response to the write operation and prior to committing the data to a data block of the primary storage, the data being deduplicated in-band in the read-write path between the virtual machine and the primary non-volatile storage while providing access to the data block in the primary storage during the deduplicating and in response to the read operation.

2. The method of claim 1, wherein deduplicating the data in the cache memory of the deduplication application further comprises performing a lookup operation of a hash in a memory index, the hash being associated with a buffer generated during the deduplicating.

3. The method of claim 1, wherein deduplicating the data in the cache memory of the deduplication application further comprises in response to the write operation, generating a data structure associated with a buffer head, and generating another data structure associated with a buffer.

4. The method of claim 3, further comprising generating a hash associated with the buffer.

5. The method of claim 4, further comprising performing a lookup operation associated with the hash, the lookup operation being configured to determine whether the hash is found in a memory index.

6. The method of claim 5, wherein if the hash is found in the memory index, performing a lookup of a reference to a parent data block and modifying an inode associated with the data to point to the parent data block and incrementing a reference counter associated with the parent data block.

7. The method of claim 5, wherein if the hash is not found in the memory index, inspecting the buffer head to determine if the operation is associated with writing a new data block or modifying an existing data block.

8. The method of claim 7, wherein if the operation is associated with the writing the new data block, modifying another inode associated with the new data block to refer to the new data block and establishing another reference counter with a value of “

1.”

9. The method of claim 7, wherein if the operation is associated with the modifying the existing data block, generating another hash associated with the existing data block and performing a copy on write operation, decrementing a further reference counter associated with the existing data block, allocating another new data block and another new data block reference counter, and modifying another inode associated with the new data block.

10. The method of claim 4, wherein the data block is read into the cache memory of the deduplication application and the write operation is performed on the data block to generate an updated data block, the updated data block being written to the primary storage asynchronously.

11. The method of claim 1, further comprising determining whether the write operation is associated with a data block of less than 4 kilobytes.

12. A system, comprising:

a cache memory of a deduplication application configured to store data received from a plurality of virtual machines and indicating a write operation, wherein the cache memory of the deduplication application is deployed in a read-write path of a virtualized computing environment between the plurality of virtual machines and a primary storage of the virtualized computing environment; and

a processor to execute the deduplication application configured

to receive a write operation and data associated with the write operation from a virtual machine, and to receive a read operation and an indication of data requested with the read operation from a second virtual machine, wherein the write operation and data are received by the deduplication application in the read-write path prior to storage of the data in a primary storage of the virtualized computing environment,

to receive the data into the cache memory of the deduplication application, the data being received from the deduplication application and indicating the write operation of the virtual machine, and

to deduplicate the data in the cache memory of the deduplication application in substantially real-time in response to the write operation and prior to committing the data to the primary storage, the data being deduplicated in-band in the read-write path between the virtual machine and the primary storage while access to the data block in the primary storage is provided during the deduplication and in response to the read operation.

13. The system of claim 12, wherein the processor further comprises the deduplication application configured to deduplicate the data inline in a data path between the primary non-volatile storage of the virtualized computing environment and a virtualization software abstraction layer.

14. The system of claim 12, wherein the processor further comprises the deduplication application configured to deduplicate the data inline in a data path between the primary storage of the virtualized computing environment and a hypervisor.

15. The system of claim 12, wherein the data is associated with a virtual machine file.

16. The system of claim 12, wherein the data is associated with one or more virtual machine files, the data being deduplicated in real time as changes are made by the virtual machine to at least one of the one or more of virtual machines files.

17. A non-transitory computer program product embodied in a computer readable medium and comprising computer instructions for:

receiving, at a deduplication application in a read-write path of a virtualized computing environment between a plurality of virtual machines and a primary storage of the virtualized computing environment, a write operation and data associated with the write operation from the virtual machine, and a read operation and an indication of data requested with the read operation from a second virtual machine, wherein the write operation and data are received by the deduplication application in the read-write path prior to storage of the data in the primary storage of the virtualized computing environment;

receiving data into a cache memory of the deduplication application, the data being received from the deduplication application and indicating the write operation of the virtual machine, wherein the cache memory of the deduplication application is deployed in the read-write path of a virtualized computing environment; and

deduplicating the data in the cache memory by the deduplication application in substantially real-time in response to the write operation and prior to committing the data to the primary storage, the data being deduplicated in-band in the read-write path between the virtual machine and the primary storage while providing access to the data block in the primary storage during the deduplication and in response to the read operation.