System and method for managing virtual hard disks in cloud environments

• US 10,044,806 B2
• Filed: 03/03/2017
• Issued: 08/07/2018
• Est. Priority Date: 02/22/2012
• Status: Active Grant

• Current Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Original Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Inventors: Yellapragada, Subrahmanya Sarma
• Primary Examiner(s): Book, Phyllis A
• Application Number:
  US15/449,299
• Publication Number:
  US 20170180479A1
• Time in Patent Office: 522 Days
• Field of Search: None
• CPC Class Codes: G06F 16/188 Virtual file systems
• CPC Class Codes: G06F 3/0605 by facilitating the interaction with a user or administrator
• CPC Class Codes: G06F 3/0631 by allocating resources to storage systems
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0643 Management of files
• CPC Class Codes: G06F 3/0664 at device level, e.g. emulation of a storage device or system
• CPC Class Codes: G06F 3/0665 at area level, e.g. provisioning of virtual or logical volumes
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• 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
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]

Abstract

A system, method, and computer-readable storage medium for managing virtual hard disks in a cloud computing/storage environment are provided. The method includes associating, using a virtual hard disk (VHD) management system of a server device, a plurality of data blocks of a virtual hard disk stored at a cloud vendor to a corresponding plurality of cloud objects. A plurality of cloud object identifiers associated with the plurality of cloud objects in a first cloud allocation table are stored. Changes to one or more data blocks are determined. Corresponding new cloud allocation tables for every data block in the plurality of data blocks that has changed are formed, the new cloud allocation tables having corresponding new cloud object identifiers. The first and the new cloud allocation tables are downloaded and merged to form an updated cloud allocation table. The updated cloud allocation table is uploaded to the cloud vendor.

Claims

1. A method, comprising:

modifying a first virtual hard disk file, having a block allocation table, to form a second virtual hard disk having a cloud allocation table, wherein;

the first virtual hard disk file comprises a disk header, a disk footer, and a plurality of data blocks storing the contents of the first virtual hard disk,

the block allocation table comprises a table of absolute sector offsets into the file, the respective offsets corresponding to respective ones of the data blocks,

the cloud allocation table comprises pointers to the data in the data blocks stored as distributed cloud objects, at least some of the distributed cloud objects being distributed in different files or blobs from one another;

creating a plurality of recovery points by, for each of at least some of the plurality of recovery points;

accessing a previous recovery point; and

for cloud objects storing data that has changed since the previous recovery point, causing the changed data to be stored in a new cloud object identified in a cloud allocation table of a current recovery point;

merging, with one or more computing devices, at least some of the recovery points without downloading to the one or more computing devices at least some of the cloud objects pointed to in respective cloud allocation tables of the merged recovery points; and

storing the merged recovery points.

2. The method of claim 1, comprising:

before merging at least some of the recovery points, determining that a threshold number of recover points have been created, wherein;

merging is performed in response to the determination, and

merging comprises merging two consecutive recovery points in a sequence of recover points.

3. The method of claim 1, wherein:

the first virtual hard disk is a dynamic virtual hard disk configured to be dynamic increased in size by adding data blocks, the dynamic virtual hard disk comprising a sector map including a bitmap that indicates which sectors contain valid data and which sectors have not been modified.

4. The method of claim 1, wherein creating the plurality of recovery points comprises:

obtaining a parent virtual hard disk corresponding to one recovery point among the plurality of recovery points, the parent virtual hard disk comprising a parent cloud allocation table comprising a pointer to a given cloud object;

determining that, in a snapshot of the first virtual hard disk for a subsequent recovery point to the one recover point, data has changed relative to a version of the data in the given cloud object; and

forming a child virtual hard disk corresponding to the subsequent recovery point, the child virtual hard disk comprising a child cloud allocation table that;

omits the pointer to the given cloud object; and

includes a pointer to a new cloud object absent from the parent cloud allocation table in a position corresponding to that of the omitted pointer.

5. The method claim 4, wherein merging the at least some of the recovery points comprises:

forming a merged virtual hard disk having a merged cloud allocation table comprising both the pointer to the new cloud object and another pointer to another cloud object present in the parent cloud allocation table.

6. The method claim 1, wherein:

the pointers in the cloud allocation table point via cloud object identifiers; and

there is a one to one mapping between cloud object identifiers and at least a plurality of the plurality of the data blocks.

7. The method of claim 1, wherein the cloud objects are data blobs.

8. The method of claim 1, wherein the cloud objects are files.

9. The method of claim 1, wherein:

a given plurality of the cloud objects are stored in a single file; and

the pointers pointing to the given plurality of the cloud objects include both a filename and an offset.

10. The method of claim 1, comprising:

restoring the first virtual hard disk to a snapshot of the first virtual hard disk with at least some of the plurality of recovery points.

11. The method of claim 10, wherein:

at least a plurality of the data blocks of the restored first virtual hard disk are contiguously stored in contiguous memory locations;

the restored first virtual hard disk comprises a restored block allocation table having pointers to the contiguous memory locations;

there is a one to one mapping between cloud object identifiers serving as pointers in cloud allocation tables of the at least some of the plurality of the recovery points and the data blocks of the restored first virtual hard disk; and

merging comprises merging without downloading to the one or more computing devices any of the cloud objects identified in respective cloud allocation tables of the merged recovery points.

12. The method of claim 1, wherein creating a plurality of recovery points comprises:

obtaining a plurality snapshots of the first virtual hard disk over time;

for each of the snapshots forming a recovery point with operations comprising;

creating a template cloud allocation table that copies a previous cloud allocation table;

determining which data blocks of the first virtual hard disk have changed since a previous snapshot or, for a first snapshot, since a base version of the first virtual hard disk;

modifying the template cloud allocation table to omit pointers to cloud objects storing the data blocks determined to have changed and add pointers to cloud objects storing current versions of the data blocks that have changed;

indicating which entries in the modified template cloud allocation table are presently modified; and

causing the modified template cloud allocation table to be stored.

13. The method of claim 12, wherein forming a recovery point comprises:

forming a sector map indicating sectors of the first virtual hard disk determined to have changed, each entry in the sector map being associated with one of the pointers to the cloud objects in the modified template cloud allocation table.

14. The method of claim 1, wherein forming a recovery point comprises:

forming a recovery-point header comprising;

a validation data structure to determine a checksum of the recovery-point header;

identifiers of a plurality of other recovery points, each corresponding to a different virtual hard disk image; and

disk information identifiers that include information related to most recent child disks in reverse chronological order.

15. The method of claim 1, wherein merging comprises:

steps for merging recovery points.

16. The method of claim 1, the operations comprising:

steps for managing virtual hard disks.

17. The method of claim 1, comprising:

providing a cloud computing or storage environment in which the cloud objects are stored; and

storing the cloud objects.

18. A tangible, non-transitory, machine-readable medium storing instructions that when executed by one or more processors effectuate operations comprising:

modifying a first virtual hard disk file, having a block allocation table, to form a second virtual hard disk having a cloud allocation table, wherein;

the first virtual hard disk file comprises a disk header, a disk footer, and a plurality of data blocks storing the contents of the first virtual hard disk,

the block allocation table comprises a table of absolute sector offsets into the file, the respective offsets corresponding to respective ones of the data blocks,

the cloud allocation table comprises pointers to the data in the data blocks stored as distributed cloud objects, at least some of the distributed cloud objects being distributed in different files or blobs from one another;

creating a plurality of recovery points by, for each of at least some of the plurality of recovery points;

accessing a previous recovery point; and

for cloud objects storing data that has changed since the previous recovery point, causing the changed data to be stored in a new cloud object identified in a cloud allocation table of a current recovery point;

merging, with one or more computing devices, at least some of the recovery points without downloading to the one or more computing devices at least some of the cloud objects pointed to in respective cloud allocation tables of the merged recovery points; and

storing the merged recovery points.

19. The medium of claim 18, the operations comprising:

restoring the first virtual hard disk to a snapshot of the first virtual hard disk with at least some of the plurality of recovery points, wherein;

at least a plurality of the data blocks of the restored first virtual hard disk are contiguously stored in contiguous memory locations,

the restored first virtual hard disk comprises a restored block allocation table having pointers to the contiguous memory locations, and

there is a one to one mapping between cloud object identifiers serving as pointers in cloud allocation tables of the at least some of the plurality of the recovery points and the data blocks of the restored first virtual hard disk.

20. The medium of claim 18, wherein creating a plurality of recovery points comprises:

obtaining a plurality snapshots of the first virtual hard disk over time; and

for each of the snapshots forming a recovery point with operations comprising;

creating a template cloud allocation table that copies a previous cloud allocation table; and

determining which data blocks of the first virtual hard disk have changed since a previous snapshot or, for a first snapshot, since a base version of the first virtual hard disk.