Copy-on-read process in disaster recovery

• US 9,892,008 B2
• Filed: 06/10/2015
• Issued: 02/13/2018
• Est. Priority Date: 07/03/2014
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Wilkinson, John P.
• Primary Examiner(s): Bryan, Jason
• Application Number:
  US14/735,886
• Publication Number:
  US 20160004602A1
• Time in Patent Office: 979 Days
• CPC Class Codes: G06F 11/2064 while ensuring consistency
• CPC Class Codes: G06F 11/2074 Asynchronous techniques
• CPC Class Codes: G06F 11/2082 Data synchronisation

Abstract

Systems, methods, and computer products for copy-on-read processes in disaster recovery include: making a disaster recovery storage volume available for read access before all data from a corresponding primary storage volume has been copied to a disaster recovery storage volume; maintaining a record of regions of the disaster recovery storage volume; in response to receiving a read request for data at the disaster recovery system: looking up the record of regions of the disaster recovery storage volume to determine available data for the read request; reading any available data from the disaster recovery storage volume; obtaining data unavailable at the disaster recovery storage volume from the corresponding primary storage volume; updating the disaster recovery storage volume with the obtained data; supplying the obtained data to the read request; and updating the record of regions of the disaster recovery storage volume for the regions of the obtained data.

Claims

1. A method for copy-on-read process in disaster recovery, comprising:

making a disaster recovery storage volume available at a disaster recovery system for read access before all data from a corresponding primary storage volume has been copied to the disaster recovery storage volume;

maintaining a record of regions of the disaster recovery storage volume, wherein a region is marked if a copy of data is available on the disaster recovery storage volume;

in response to receiving a read request for data at the disaster recovery system;

looking up the record of regions of the disaster recovery storage volume to determine available data for the read request;

reading any available data from the disaster recovery storage volume;

for any data unavailable at the disaster recovery storage volume, obtaining, the unavailable data from the corresponding primary storage volume;

updating the disaster recovery storage volume with the obtained data;

supplying the obtained data to the read request; and

updating the record of regions of the disaster recovery storage volume for the regions of the obtained data;

receiving an update for a region of the disaster recovery storage volume from a primary system; and

in response to determining no marking exists in the record of regions for the region of the disaster recovery storage volume, discarding the update, and

wherein the marking indicates that a copy of data is available on the disaster recovery storage volume.

2. The method as recited in claim 1, comprising:

maintaining a record of regions of the corresponding primary storage volume wherein a region is marked when initial data is sent to a disaster recovery system; and

using the record of regions of the corresponding primary storage volume to determine whether later writes should be immediately replicated to the disaster recovery system.

3. The method as recited in claim 1, wherein a region in the record of regions corresponds in size to either:

a minimum write granularity of attached host systems;

or

an amount of data larger than the minimum write granularity.

4. The method as recited in claim 1, wherein looking up the record of regions further comprises:

looking up one or more markings in the record of regions, the markings corresponding to one or more logical block addresses in a range of the read request; and

reading any available data from the disaster recovery storage volume for one or more available logical block addresses.

5. The method as recited in claim 1, wherein obtaining any unavailable data from the corresponding primary storage volume comprises obtaining a range of data comprising an integer number of regions corresponding to a range of the read request.

6. The method as recited in claim 1, wherein supplying the obtained data to the read request comprises satisfying the read request from the disaster recovery storage volume or from copied data in system memory.

7. The method as recited in claim 1, wherein the record of regions is a bitmap with each bit corresponding to a region of the disaster recovery storage volume.

8. The method as recited in claim 1, comprising:

determining if overlapping read requests are received at the disaster recovery system; and

in response to determining one or more overlapping read requests were received at the disaster recovery system, selectively obtaining only one copy of data from a primary system for the one or more overlapping read requests.

9. The method as recited in claim 1, comprising:

determining if there is an interruption during obtaining data from the primary system; and

resubmitting a copy request immediately or failing the read request.

10. The method as recited in claim 1, comprising:

prioritizing one or more uncopied regions to be obtained from the primary system.

11. A system for copy-on-read process in disaster recovery, comprising:

a disaster recovery storage volume comprising a computer readable storage medium available at a disaster recovery system for read access before all data from a corresponding primary storage volume has been copied to the disaster recovery storage volume;

a record of regions of the disaster recovery storage volume, wherein a region is marked if a copy of data is available on the disaster recovery storage volume; and

a disaster recovery replication component configured to receive a read request for data at the disaster recovery system, the disaster recovery replication component further comprising;

a disaster recovery record look up component configured to look up the record of regions to determine available data for the read request;

a disaster recovery read component configured to read any available data from the disaster recovery storage volume; and

a primary copy component further comprising;

a primary message component configured to obtain any unavailable data from the corresponding primary storage volume,

a disaster recovery write component configured to update the disaster recovery storage volume with the obtained data, a host reply component configured to supply the obtained data to the read request, and

a record mark up component configured to update the record of regions for the regions of the obtained data; and

an update component configured to;

receive an update for a region of the disaster recovery storage volume from a primary system; and

in response to determining no marking for the region exists in the record of regions, discarding the update, and

wherein the marking indicates that a copy of data is available on the disaster recovery storage volume.

12. The system as recited in claim 11, wherein the host reply component satisfies the read request from either:

the disaster recovery storage volume;

or

copied data in system memory.

13. The system as recited in claim 11, wherein the record of regions is a B-tree.

14. The system as recited in claim 11, comprising an overlapping read component configured to:

determine if overlapping read requests are received at the disaster recovery system; and

in response to determining one or more overlapping read requests were received at the disaster recovery system, selectively obtain only one copy of data from the primary system for the overlapping read requests.

15. The system as recited in claim 11, comprising an interruption component configured to:

determine if an interruption occurs during obtaining data from the primary system; and

resubmit a copy request either immediately or upon failing the read request, in response to determining the interruption occurred.

16. The system as recited in claim 11, comprising a prioritizing component configured to prioritize uncopied regions obtained from the primary system.

17. The system as recited in claim 11, comprising a primary system, the primary system comprising:

a record of regions of the corresponding primary storage volume configured to mark a region when initial data is sent to a disaster recovery system; and

a record monitoring component configured to use the record of regions of the corresponding primary storage volume to determine whether one or more later writes should be immediately replicated to the disaster recovery system.

18. The system as recited in claim 11, wherein a region in the record of regions corresponds in size to an amount of data larger than a minimum write granularity of one or more attached host systems.

19. The system as recited in claim 11, wherein the disaster recovery record look up component is further configured to:

look up a marking in the record or regions for one or more logical block addresses in a range of the read request; and

read any available data from the disaster recovery storage volume for one or more available logical block addresses.

20. A computer program product for copy-on-read process in disaster recovery, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions readable/executable by a computer to cause the computer to perform a method comprising:

make, by the computer, a disaster recovery storage volume available at a disaster recovery system for read access before all data from a corresponding primary storage volume has been copied to the disaster recovery storage volume;

maintain, by the computer, a record of regions of the disaster recovery storage volume wherein a region is marked if a copy of data is available on the disaster recovery storage volume;

in response to receipt of a read request for data at the disaster recovery system;

look up the record of regions to determine available data for the read request;

read any available data from the disaster recovery storage volume;

obtain, from the corresponding primary storage volume, any data unavailable at the disaster recovery storage volume;

update the disaster recovery storage volume with the obtained data;

supply the obtained data to the read request; and

update the record of regions of the disaster recovery storage volume for the regions of the obtained data; and

receive an update for a region of the disaster recovery storage volume from a primary system; and

in response to determining no marking exists in the record of regions for the region of the disaster recovery storage volume, discard the update, and

wherein the marking indicates that a copy of data is available on the disaster recovery storage volume.

21. The computer program product as recited in claim 20, wherein the program instructions configured to cause the computer to look up the record of regions comprises program instructions configured to cause the computer to:

look up one or more markings in the record of regions, the one or more markings corresponding to one or more logical block addresses in a range of the read request; and

read any available data from the disaster recovery storage volume for one or more available logical block addresses.