Method for Remote Asynchronous Replication of Volumes and Apparatus Therefor

• US 20100205392A1
• Filed: 01/22/2010
• Published: 08/12/2010
• Est. Priority Date: 01/23/2009
• Status: Active Grant

• Current Assignee: Infortrend Technology Incorporated (Infortrend Corporation)
• Original Assignee: Infortrend Technology Incorporated (Infortrend Corporation)
• Inventors: Schnapp, Michael Gordon, Fang, Ching-Hua
• Granted Patent: US 9,569,321 B2
• Time in Patent Office: Days
• US Class Current: 711/162
• CPC Class Codes: G06F 11/2066 Optimisation of the communication load
• CPC Class Codes: G06F 11/2069 Management of state, configuration or failover
• CPC Class Codes: G06F 11/2074 Asynchronous techniques
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/065 Replication mechanisms
• 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]

Abstract

A method for remote asynchronous volume replication and apparatus therefore are disclosed. Asynchronous replication is applied to deal with data changes on the source volume on the local site incurred by Host IO requests. In coordination with the “point-in-time differential backup” technology, the data is subjected to be backuped to Source BAS on the local site (backup-on-write operation) only when the original data being written into the block of the source volume is different from the data of the corresponding block of the destination volume on the remote site. As a result, once a new data is written into the source volume completely, the host will be responded that its Host IO request is completed. Therefore, the data necessarily transmitted to the destination volume on the remote site can be minimized, and the problem of remote data transmission limited by network bandwidth can be prevented effectively, thereby keeping the operation performance of the storage system at a better level.

Claims

1. A method for asynchronous replication of volumes, comprising:

mirroring a source volume located on a first site to a destination volume located on a second site;

receiving a Host input/output (IO) request, wherein the Host IO request is sent from a host located on the first site for writing an updated data set into a first data block of the source volume;

executing a backup-on-write operation, wherein the backup-on-write operation firstly determines whether an original data set of the first data block of the source volume is the same with a data set of a second data block, corresponding to the first data block, of the destination volume, so as to determine whether the original data set of the first data block of the source volume is backuped to a first backup image located on the first site or not;

writing the updated data set into the first data block of the source volume; and

responding the host that the Host IO request has been completed.

2. The method of claim 1, further comprising:

performing a copy procedure for copying the data stored in the first backup image or stored in the source volume into the destination volume.

3. The method of claim 2, wherein the copy procedure uses a communication medium to transmit the data from the first site to the second site.

4. The method of claim 1, further comprising:

allocating a first entry in a volume log table (VLT) of the source volume, and initializing a plurality of fields of the first entry comprising;

setting a sequence number (SeqNum) defined in a first field to an initial setting value; and

setting an operation code (OpCod) defined in a second field to an operation code representing a state of the source volume;

wherein the initial setting value recorded in the first field is further assigned to a first variable;

allocating a second entry in a VLT of the destination volume, and initializing a plurality of fields of the second entry comprising;

setting an OpCod defined in a second field of the second entry to an operation code representing a state of the destination volume;

adding an increment to a value of a master sequence number variable of the destination volume, and copying the value of the master sequence number variable of the destination volume into a first field of the second entry in the VLT of the destination volume as a sequence number defined therein, and into a fourth field of the first entry in the VLT of the source volume as an alternative volume'"'"'s sequence number (AltVolSN) defined therein, wherein the value of the master sequence number variable of the destination volume is further assigned to a second variable;

setting a fourth field of the second entry in the VLT of the destination volume as a starting address, wherein the fourth field of the second entry is used for recording an index of a next data block to be synchronously replicated (DstVLT.NxtSyn Blk);

adding an increment to a value of a master sequence number variable of the source volume, and copying the value of the master sequence number variable of the source volume into a first field of the first entry in the VLT of the source volume as a sequence number defined therein, wherein the value of the master sequence number variable of the source volume is further assigned to a first variable;

allocating a third entry in the VLT of the destination volume, and initializing a plurality of fields of the third entry comprising;

setting an OpCod defined in a second field of the third entry to an operation code representing a split state;

adding an increment to the value of the master sequence number variable of the destination volume again, and copying the value of the master sequence number variable of the destination volume into a first field of the third entry in the VLT of the destination volume as a sequence number defined therein; and

determining if the synchronization operation between the source volume and the destination volume is complete.

5. The method of claim 4, further comprising:

performing a background copy procedure for copying the source volume to the destination volume.

6. The method of claim 4, further comprising:

determining if there is a host IO (Input/Output) request targeting the source volume for data access.

7. The method of claim 6, further comprising:

transforming the host IO request into at least one logical drive (LD) IO request, wherein each of the at least one LD IO request is targeting a data block, which is to be accessed, of the source volume.

8. The method of claim 7, wherein, for each of the at least one LD IO request, the method further comprises:

determining if the host IO request is a write request;

executing a procedure of checking/copying data blocks from the source volume to the destination volume (Check/Copy Block from Source to Destination);

writing data of the host IO request to the data block, which is to be accessed, of the source volume; and

setting a sequence number stored in an entry of a volume block update table (VBUT) of the source volume to be equal to the value of the first variable, in which the entry is associated with the to-be-accessed data block.

9. The method of claim 7, further comprising:

determining if the host IO request is a write request;

determining if the host IO request is targeting the source volume; and

reading data from the data block, which is to be accessed, of the source volume.

10. The method of claim 7, further comprising:

determining if the host IO request is a write request;

determining if the host IO request is targeting the source volume;

determining if the host IO request is targeting the destination volume and determining if a sequence number (DstVBUT.SeqNum) of the entry, which is currently processed, in the VBUT of the destination volume is less than a value of the second variable; and

reading data from the data block to be accessed in the source volume or from the corresponding data block in the first backup image.

11. The method of claim 7, further comprising:

determining if the host IO request is a write request;

determining if the host IO request is targeting the source volume;

determining if the host IO request is targeting the destination volume and determining if a sequence number (DstVBUT.SeqNum) of the entry, which is currently processed, in the VBUT of the destination volume is less than the value of the second variable; and

reading data from the data block to be accessed in the destination volume.

12. The method of claim 7, further comprising:

determining whether at least one procedure respectively associated with the at least one LD IO request is all complete or not.

13. The method of claim 12, further comprising:

responding a completion message of the host IO request to a host issuing the host IO request.

14. The method of claim 5, wherein the background copy procedure for copying the source volume to the destination volume comprises:

for each data block associated with each entry in a VBUT of the destination volume, calling and executing a procedure of checking/copying data blocks from the source volume to the destination (Check/Copy Block from Source to Destination); and

determining if any data block associated with the VBUT of the destination volume is not being synchronized yet.

15. The method of claim 14, further comprising:

adding the DstVLT.NxtSynBlk recorded in the fourth field of the second entry in the VLT of the destination volume by 1 unit if it is determined that there is a data block associated with the VBUT of the destination volume left to be synchronized.

16. The method of claim 14, further comprising:

setting the DstVLT.NxtSynBlk recorded in the fourth field of the second entry in the VLT of the destination volume to a predetermined value used for representing that all of the data blocks have been synchronously replicated if it is determined that all of the data blocks associated with the VBUT of the destination volume are already synchronized.

17. The method of claim 16, wherein the predetermined value is “

−

1”

.

18. The method of claim 8, wherein the procedure of Check/Copy Block from Source to Destination comprises:

determining if a sequence number (DstVBUT.SeqNum) of an entry in a VBUT of the destination volume is less than the value of the second variable, in which the entry is associated with a data block being processed.

19. The method of claim 18, further comprising:

executing a procedure of checking whether a volume block is in synchronization (Check Volume Block In-Sync), and checking if the procedure of Check Volume Block In-Sync returns a value of “

FALSE”

.

20. The method of claim 19, further comprising:

copying data of the source volume to the first backup image as a backup-on-write operation.

21. The method of claim 14, wherein the procedure of Check/Copy Block from Source to Destination comprises:

determining if a sequence number (DstVBUT.SeqNum) of an entry in a VBUT of the destination volume is less than the value of the second variable, in which the entry is associated with a data block being processed.

22. The method of claim 21, further comprising:

executing a procedure of checking whether a volume block is in synchronization (Check Volume Block In-Sync), and checking if the procedure of Check Volume Block In-Sync returns a value of “

FALSE”

.

23. The method of claim 22, further comprising:

determining if the destination volume is accessible.

24. The method of claim 23, further comprising:

setting a synchronization error flag (SyncErrorFlag) to “

TRUE”

.

25. The method of claim 23, further comprising:

determining if a corresponding data block of the first backup image is empty.

26. The method of claim 25, further comprising:

copying the data stored in the corresponding data block of the first backup image to the destination volume; and

setting the DstVBUT.SeqNum to the value of the second variable.

27. The method of claim 25, further comprising:

determining if a sequence number (SrcVBUT.SeqNum) of an entry in a VBUT of the source volume is less than the first variable, in which the entry is associated with a data block being processed; and

setting the DstVBUT.SeqNum to the value of the second variable.

28. The method of claim 25, further comprising:

determining if a sequence number (SrcVBUT.SeqNum) of an entry in a VBUT of the source volume is less than the first variable, in which the entry is associated with a data block being processed;

copying the data stored in the data block of the source volume to the destination volume; and

setting the DstVBUT.SeqNum to the value of the second variable.

29. The method of claim 19 or claim 22, wherein the procedure of Check Volume Block In-Sync comprises five determining steps of:

determining if the DstVBUT.SeqNum is equal to an initial setting value;

determining if an entry in the VLT of the destination volume, in which the entry is associated with the DstVBUT.SeqNum, is missing, or can be found but the data cannot be read, or the data are inconsistent;

determining if the operation code (DstVLT.OpCod) recorded in the second field of the entry, which is being processed, in the VLT of the destination volume is not the operation code representing the state of the destination volume;

or if an alternative volume'"'"'s identification (DstVLT.AltVolID) recorded in a third field of the entry, which is currently processed, in the VLT of the destination volume is not an identification code of the source volume corresponding thereto;

determining if an entry in the VLT of the source volume having an alternative volume'"'"'s sequence number (SrcVLT.AltVolSN), which is equal to the DstVBUT.SeqNum, recorded in the fourth field of the entry, which is being processed, is missing, or can be found but the data cannot be read or the data are not consistent; and

determining if the SrcVBUT.SeqNum is greater than or equal to a sequence number (SrcVLT.SeqNum) recorded in the first field of the entry, which is being processed, in the VLT of the source volume.

30. The method of claim 29, further comprising:

responding the “

FALSE”

message if any of the determination results of the five determining steps is “

yes”

.

31. The method of claim 29, further comprising:

responding a “

TRUE”

message if all of the determination results of the five determining steps is “

no”

.

32. The method of claim 29, wherein the initial setting value is “

0(null)”

.

33. The method of claim 4, further comprising an asynchronous replication failover process, which comprises:

unmapping the destination volume and a second section allocation system (SAS) of the destination volume from a first storage virtualization subsystem located on the first site;

mounting the second SAS and a second backup image on a second controller of a second storage virtualization subsystem on the second site, in which the second backup image is stored on the second site;

initializing a rollback operation for recovering data of the destination volume to the state recorded by the second backup image, and using the VBUT of the destination volume to keep tracking and recording data changes caused by the rollback operation;

splitting the destination volume from the second backup image; and

mapping the destination volume to the host for serving a request from the host, and using the VBUT of the destination volume to keep tracking and recording data changes caused by the host.

34. The method of claim 33, further comprising an asynchronous replication failback process, which comprises:

unmapping the destination volume from the host and dismounting the second SAS of the destination volume;

mapping the destination volume and the second SAS to a first controller of a first storage virtualization subsystem located on the first site;

mounting a first SAS of the source volume and the second SAS of the destination volume on the first controller;

initializing a synchronous resynchronization operation for replicating data of the destination volume to the source volume, and splitting the destination volume from the source volume;

mapping the source volume to the host; and

changing a replication mode between the source volume and the destination volume to be asynchronous when the synchronous resynchronization operation is complete.

35. The method of claim 33, further comprising an asynchronous replication failback process, which comprises:

mapping the source volume and a first SAS of the source volume to the second controller of the second storage virtualization subsystem located on the second site;

mounting the first SAS of the source volume on the second controller located on the second site; and

initializing a synchronous or asynchronous resynchronization operation for synchronously or asynchronously replicating data of the destination volume to the source volume.

36. The method of claim 35, further comprising:

unmapping the destination volume from the host;

dismounting the second SAS of the destination volume and the first SAS of the source volume from the second controller on the second site;

unmapping the source volume and the first SAS from the second controller located on the second site;

mapping the destination volume and the second SAS to the first controller located on the first site;

mounting the first SAS, the second SAS and the first backup image on the first controller located on the first site;

initializing a rollback operation for recovering data of the source volume to the state recorded by the first backup image, and using the VBUT of the source volume to keep tracking and recording data changes caused by the rollback operation;

splitting the source volume from the first backup image;

mapping the source volume to the host and using the VBUT of the destination volume to keep tracking and recording data changes caused by the host; and

performing an asynchronous replication operation.

37. A system for asynchronous volume replication, comprising:

a first storage virtualization subsystem located on a first site, comprising a first controller and a first storage unit, wherein the first storage unit has a first volume and a first backup image;

a second storage unit located on a second site, wherein the second storage unit has a second volume; and

a host located on the first site and coupled with the first storage virtualization subsystem, wherein the host sends a Host IO request to the first volume of the first storage virtualization subsystem;

wherein when the first controller of the first storage virtualization subsystem receives the Host IO request, a backup-on-write operation is executed to determine whether an original data set of a first data block, which is to be written, of the first volume is the same with a data set of a second data block, corresponding to the first data block, of the second volume, so as to determine whether the original data of the first data block of the first volume is backuped to the first backup image or not;

writing an updated data of the Host IO request into the first data block of the first volume by the first controller; and

responding the host that the Host IO request is completed.

38. The system of claim 37, wherein the second storage unit further comprises a second backup image for recording original data of the second volume when the data of the second volume changes.

39. The system of claim 37, wherein a communication medium is used to transmit data between the first storage virtualization subsystem located on the first site and the second storage virtualization subsystem located on the second site.

40. The system of claim 37, wherein the communication medium is selected from the group consisting of a Local Area Network (LAN), a Wide Area Network (WAN), a Storage Area Network (SAN), Internet, and a Dedicated Data Transfer Channel.

41. The system of claim 37, wherein the first backup image is administrated by a section allocation system (SAS).

42. The system of claim 37, wherein the section allocation system (SAS) comprising a first attribute block used for recording an identification (ID) of the SAS and at least one ID of at least one media extent associated thereto, a block association set (BAS) related field used for recording information pointing a block association set (BAS) as well as information of the source volume associated with the BAS, and a first journal field used for storing an operating journal so as to provide a recovery while an error or power failure occurs in the system.

43. The system of claim 42, wherein the BAS related field further comprises a section allocation table (SAT) and an inode table.

44. The system of claim 42, wherein the BAS further comprises a second attribute block used for recording an ID of the BAS, a size of the BAS and other relative setting of the BAS, a directory field used for recording a level of a block association table (BAT) and a quantity of the BAT(s) that has(have) been established, a second journal field used for storing an operation journal so as to provide a recovery while an error or power failure occurs in the system, a folder table used for pointing the BAT, and a data area used for storing the first backup image.