IP Library Granted Patent US 9,326,001
Granted Patent B2
US 9,326,001 · App. 13/053,489 · Granted Apr 26, 2016

Scalable image distribution in virtualized server environments

Inventors: Michael R. Head (Tarrytown, NY); Andrzej Kochut (Croton-On-Hudson, NY); Hidayatullah H. Shaikh (Shrub Oak, NY)
Assignee: International Business Machines Corporation
H04N19/507
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,326,001
App. No.
13/053,489
Granted
Apr 26, 2016
Kind
B2
Abstract

A method and system include replicating an image representing a sequence of bytes on a local storage medium on a target device by determining a similarity between images and reconstructing a new image using equivalent blocks from one or more similar images locally available on the target device or available on donor devices to reduce network link usage and transfer time in replicating the image.

Claims (29)

1. A method, comprising:

replicating an image representing a sequence of bytes on a local storage medium on a target device by:

determining a similarity between images; and

reconstructing a new image by copying data from equivalent blocks from one or more similar images locally available on the target device or available on a best fit donor devices if no similar images are locally available, to reduce network link usage and transfer time in replicating the image, wherein the new image represents a virtual machine (VM).

2. The method as recited in claim 1 , wherein determining a similarity includes performing a checksum on a byte-by-byte comparison between two blocks of images with a same hashcode.

3. The method as recited in claim 1 , wherein the similarity is stored in an image dependency matrix denoting a fraction of blocks that are identical between two images.

4. The method as recited in claim 3 , wherein the image dependency matrix includes an index for identical blocks between the two images.

5. The method as recited in claim 1 , wherein reconstructing includes querying donor devices for the equivalent blocks which are not available in local storage but are needed to reconstruct the image.

6. The method as recited in claim 1 , wherein reconstructing includes accessing a storage server to find remaining blocks needed to reconstruct the new image.

7. The method as recited in claim 1 , wherein reconstructing the new image includes creating a copy-on-write copy of the one or more similar images.

8. The method as recited in claim 1 , wherein the best fit donor device for reconstructing the new image is determined by evaluating at least one criteria for a plurality of donors from a set of qualified donors.

9. The method as recited in claim 8 , wherein the criteria includes at least one of bandwidth, utilization, security, latency, or cost.

10. The method as recited in claim 8 , wherein querying includes providing an overlay network to search for donors.

11. The method as recited in claim 1 , wherein the reconstructing includes creating a new VM instance or migrating an existing VM instance.

12. The method as recited in claim 11 , further comprising starting the new VM without all of the blocks of the new VM being locally available.

13. A method, comprising:

determining a similarity between images in a distributed network;

determining an image dependency matrix denoting a fraction of blocks that are identical between pairs of images;

searching local storage for blocks to reconstruct a new image using the image dependency matrix;

if all blocks needed for the new image are not in local storage, querying donor hosts for the blocks to determine a best fit donor host for reconstructing the new image using the image dependency matrix and copying data from the fraction of blocks that are identical between pairs of images; and

if all blocks needed for the new VM are not in available from the donor hosts, accessing a storage server to find remaining blocks needed to reconstruct the new image, wherein the new image represents a virtual machine (VM).

14. The method as recited in claim 13 , wherein determining a similarity includes performing a checksum on a byte-by-byte comparison between two images with a same hashcode.

15. The method as recited in claim 13 , wherein determining an image dependency matrix includes storing an index for identical blocks between the pairs of images.

16. The method as recited in claim 13 , wherein searching local storage includes creating a copy-on-write copy of a locally available image to construct the new image.

17. The method as recited in claim 13 , wherein the best fit donor device for reconstructing the new image is determined by evaluating at least one criteria for a plurality of host donors from a set of qualified host donors.

18. The method as recited in claim 17 , wherein the criteria includes at least one of bandwidth, utilization, latency or cost.

19. The method as recited in claim 17 , wherein querying includes providing an overlay network to search for donor hosts.

20. The method as recited in claim 13 , wherein the method includes creating a new VM instance or migrating an existing VM instance.

21. The method as recited in claim 20 , further comprising starting the new VM without all of the blocks of the new VM being locally available.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 22, 2011
From: HEAD, MICHAEL R.; KOCHUT, ANDRZEJ; SHAIKH, HIDAYATULLAH H.
To: INTERNATIONAL BUSINESS MACHINES CORPORATION
Reel/Frame 025996/0102 →
Continuity (1)
Related Publication 20120243795A1 · Sep 27, 2012