Efficient VM migration across cloud using catalog aware compression

• US 10,210,011 B2
• Filed: 06/01/2016
• Issued: 02/19/2019
• Est. Priority Date: 03/26/2016
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Gautam, Ajay, Gaurav, Kumar, Chagalakondu, Gurusreekanth, Nair, Jayalekshmy, George, Jobin
• Primary Examiner(s): Bullock, Jr., Lewis A
• Assistant Examiner(s): Lu, Kevin X
• Application Number:
  US15/169,772
• Publication Number:
  US 20170277555A1
• Time in Patent Office: 993 Days
• CPC Class Codes: G06F 16/137 Hash-based content-based indexing of textual data G06F16/31
• CPC Class Codes: G06F 16/1752 based on file chunks
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 2009/45595 Network integration; Enabling network access in virtual machine instances
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: H04L 69/04 Protocols for data compression, e.g. ROHC

Abstract

Techniques for migrating a VM in a hybrid cloud computing system are provided. The techniques include identifying a similar VM at the destination, comparing dictionaries for the VM to be transmitted and the similar VM, and compressing the VM based on the comparison. After transmitting the compressed VM, the destination decompresses the VM using the dictionary of the similar VM. Dictionaries associate chunks of VM data with hashes of those chunks. This allows replacement of chunks with the hashes, thereby compressing the VM for transmission.

Claims

1. A method for migrating a first virtual machine from a source computing system to a destination computing system, the method comprising:

identifying a first dictionary for the first virtual machine;

comparing the first dictionary with a second dictionary that corresponds to a second virtual machine identified from a catalog of virtual machine templates at the destination computing system, to output a set of common hashes;

based on the set of common hashes, compressing the first virtual machine to generate a compressed virtual machine comprised of hashes in the set of common hashes without corresponding data chunks and at least one hash without corresponding data chunks, or comprised of hashes of the first dictionary without corresponding data chunks, wherein the at least one hash is found in the first dictionary but is not found in the second dictionary; and

transmitting the compressed virtual machine and a differential dictionary that comprises the at least one hash and at least one corresponding data chunk to the destination computing system.

2. The method of claim 1, wherein the first dictionary comprises:

a set of chunks of data for the first virtual machine; and

a set of hashes, each hash being a hash of a different chunk of the set of chunks of data.

3. The method of claim 2, further comprising generating the first dictionary by:

chunking a virtual machine disk file to generate the set of chunks; and

hashing the chunks of the set of chunks to generate the set of hashes.

4. The method of claim 1, wherein compressing the first virtual machine comprises:

replacing chunks of the first virtual machine with hashes of the set of common hashes.

5. The method of claim 1, wherein comparing the first dictionary with the second dictionary comprises:

identifying, as the set of common hashes, hashes within the first dictionary that are also present in the second dictionary.

6. The method of claim 1, further comprising:

obtaining first metadata that indicates software installed on the first virtual machine;

comparing the first metadata to metadata associated with virtual machines at the destination computing system; and

identifying the second virtual machine based on the comparing.

7. The method of claim 1, wherein:

an identical virtual machine to the second virtual machine exists at the source computing system, and

the source computing system does not receive the second dictionary from the destination system.

8. The method of claim 1, wherein:

an identical virtual machine to the second virtual machine does not exist at the source computing system, and

the source computing system receives a set of hashes associated with the second dictionary from the destination system.

9. The method of claim 1, further comprising:

decompressing the compressed virtual machine by replacing hashes in the compressed virtual machine with chunks from the second dictionary and the differential dictionary.

10. A system for migrating a first virtual machine from a source computing system to a destination computing system, the system comprising:

the source computing system, comprising;

a hypervisor configured to execute the first virtual machine; and

a migration module configured to;

identify a first dictionary for the first virtual machine,

compare the first dictionary with a second dictionary that corresponds to a second virtual machine identified from a catalog of virtual machine templates at the destination computing system, to output a set of common hashes,

based on the set of common hashes, compress the first virtual machine to generate a compressed virtual machine comprised of hashes in the set of common hashes without corresponding data chunks and at least one hash without corresponding data chunks, or comprised of hashes of the first dictionary without corresponding data chunks, wherein the at least one hash is found in the first dictionary but is not found in the second dictionary, and

transmit the compressed virtual machine and a differential dictionary that comprises the at least one hash and at least one corresponding data chunk to the destination computing system.

11. The system of claim 10, wherein the first dictionary comprises:

a set of chunks of data for the first virtual machine; and

a set of hashes, each hash being a hash of a different chunk of the set of chunks of data.

12. The system of claim 11, wherein the migration module is configured to generate the first dictionary by:

chunking a virtual machine disk file to generate the set of chunks; and

hashing the chunks of the set of chunks to generate the set of hashes.

13. The system of claim 10, wherein the migration module is configured to compress the first virtual machine by:

replacing chunks of the first virtual machine with hashes of the set of common hashes.

14. The system of claim 10, wherein the migration module is configured to compare the first dictionary with the second dictionary by:

identifying, as the set of common hashes, hashes within the first dictionary that are also present in the second dictionary.

15. The system of claim 10, wherein the destination system is configured to:

obtain first metadata that indicates software installed on the first virtual machine;

compare the first metadata to metadata associated with virtual machines at the destination computing system; and

identify the second virtual machine based on the comparing.

16. The system of claim 10, wherein:

an identical virtual machine to the second virtual machine exists at the source computing system, and

the source computing system does not receive the second dictionary from the destination system.

17. The system of claim 10, wherein:

an identical virtual machine to the second virtual machine does not exist at the source computing system, and

the source computing system receives a set of hashes associated with the second dictionary from the destination system.

18. The system of claim 10, wherein the destination system is configured to decompress the compressed virtual machine by replacing hashes in the compressed virtual machine with chunks from the second dictionary and the differential dictionary.

19. A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform a method for migrating a first virtual machine from a source computing system to a destination computing system, the method comprising:

identifying a first dictionary for the first virtual machine;

comparing the first dictionary with a second dictionary that corresponds to a second virtual machine identified from a catalog of virtual machine templates at the destination computing system, to output a set of common hashes;

based on the set of common hashes, compressing the first virtual machine to generate a compressed virtual machine comprised of hashes in the set of common hashes without corresponding data chunks and at least one hash without corresponding data chunks, or comprised of hashes of the first dictionary without corresponding data chunks, wherein the at least one hash is found in the first dictionary but is not found in the second dictionary; and

transmitting the compressed virtual machine and a differential dictionary that comprises the at least one hash and at least one corresponding data chunk to the destination computing system.

20. The non-transitory computer-readable medium of claim 19, wherein the first dictionary comprises:

a set of chunks of data for the first virtual machine; and

a set of hashes, each hash being a hash of a different chunk of the set of chunks of data.