Methods and apparatus for accessing data from a primary data storage system for secondary storage

• US 20060010227A1
• Filed: 09/16/2005
• Published: 01/12/2006
• Est. Priority Date: 06/01/2004
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: InMage Systems, Inc. (Microsoft Corporation)
• Inventors: Atluri, Rajeev
• Granted Patent: US 8,055,745 B2
• Time in Patent Office: Days
• US Class Current: 709/217
• CPC Class Codes: G06F 11/1453 using de-duplication of the data
• CPC Class Codes: G06F 11/1456 Hardware arrangements for backup
• CPC Class Codes: G06F 11/1464 for networked environments
• CPC Class Codes: G06F 11/1469 Backup restoration techniques
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• 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 for providing secondary data storage and recovery services for one or more networked host nodes includes a server application for facilitating data backup and recovery services; a first data storage medium accessible to the server application; a second data storage medium accessible to the server application; at least one client application for mapping write locations allocated by the first data storage medium to write locations represented in a logical view of the first data storage medium; and at least one machine instruction enabling direct read capability of the first data storage medium by the server application for purposes of subsequent time-based storage of the read data into the secondary data storage medium.

Claims

1. A system for providing secondary data storage and recovery services for one or more networked host nodes comprising:

a server application for facilitating data backup and recovery services;

a first data storage medium accessible to the server application;

a second data storage medium accessible to the server application;

at least one client application for mapping write locations allocated by the first data storage medium to write locations represented in a logical view of the first data storage medium; and

at least one machine instruction enabling direct read capability of the first data storage medium by the server application for purposes of subsequent time-based storage of the read data into the secondary data storage medium.

2. The system of claim 1 wherein the networked host nodes are connected to a local area network (LAN).

3. The system of claim 1 wherein the host nodes include but are not limited to server nodes.

4. The system of claim 2 implemented in conjunction with a storage area network (SAN) accessible through a Fibre Channel network.

5. The system of claim 2 implemented in conjunction with a SAN accessible through an Ethernet™

network.

6. The system of claim 1 wherein the second storage medium is one of a variant of a random access disk including a serial-advanced-technology-attachment (SATA) or parallel-advanced-technology-attachment (PATA) drive system.

7. The system of claim 1 wherein the at least one machine instruction for enabling direct read capability includes the write location mapping results between the first storage medium and a logical view of the first system maintained at the host.

8. The system of claim 7 wherein the logical view is a file system view of the first storage system.

9. The system of claim 1 wherein the client application includes a filter for monitoring host activity for data write activity.

10. The system of claim 1 wherein the server application includes a component for comparing new data writes read from the first storage medium against old write data previously read from the first storage medium and stored on the second storage medium, the comparison to determine whether the new data is valid change data or whether it is redundant data.

11. The system of claim 10 wherein the comparison is made against the compression weight of the payload portion of the compared writes.

12. The system of claim 1 further including a tape drive system for long term backup of data

13. The system of claim 1 wherein the server application is stored and executes from a node remote to the LAN network supporting the host nodes.

14. The system of claim 1 wherein the machine instruction references an actual offset location in the first storage medium where the original write is located.

15. A machine-readable instruction for causing an application to initiate a data-read operation against a first storage system for one or more data writes for the purpose of optionally storing one or more of the read data writes into a second storage system and preserving the versions thereof for the time of read from the first storage system for latter viewing and recovery comprising:

a portion thereof for specifying origin machine address and destination machine address of the one or more data writes;

a portion thereof for specifying one or more storage devices or partitions in the first storage system to receive the one or more data writes;

a portion thereof for specifying one or more physical memory offsets in the one or more storage devices or partitions, the offsets earmarking the beginning of the payload portion of the one or more data writes; and

a portion thereof for specifying bit length of the one or more payloads of the one or more data writes.

16. The instruction of claim 15 wherein the instruction is prepared by the host node that originally saved the write data to the first storage system.

17. The instruction of claim 15 wherein the first storage system is a SAN-based primary storage system and the second storage system is a SATA or PATA disk drive.

18. The instruction of claim 16 wherein the host node is one of a plurality of LAN-connected nodes.

19. The instruction of claim 18 wherein the host nodes include but are not limited to server nodes.

20. The instruction of claim 15 wherein the machine addresses are IP addresses.

21. The instruction of claim 15 wherein the machine addresses are of a type of machine address configurations (MAC) addresses, or worldwide node name (WWNN) addresses.

22. The instruction of claim 15 wherein the one or more storage devices or partitions are specified by logical unit numbers (LUNs).

23. The instruction of claim 15 wherein the data read operation is conducted through a Fibre Channel switch using a Host Bus Adaptor.

24. The instruction of claim 15 wherein the instruction set is generated in full or in part by running an algorithm at the point of the generating host.

25. A method for causing a data read operation against a first storage system, the data read operation performed by a server associated directly with a second storage system according to a machine-readable instruction generated by a host machine directly associated with the first storage system comprising steps of:

(a) saving data from the host machine to the first storage system;

(b) creating a record of the data write or writes comprising the data saved to the first storage system;

(c) processing, using variables contained in the record, a mapping scheme resulting in knowledge of one or more physical offsets in the first storage system to receive the one or more data writes;

(d) forwarding the record including the offset information to the server directly associated with the second storage system;

(e) accessing the first storage system to read the data according to the instructions of record; and

(f) reading the data.

26. The method of claim 25 wherein in step (a) the host machine is a LAN connected server and the first storage system is a SAN-based storage system.

27. The method of claim 26 wherein the SAN-based storage system is accessible through a Fibre Channel network, or through an Ethernet™

network.

28. The method of claim 25 wherein in step (b) the record created includes at least origin and destination machine address information, physical payload length information, logical write location information as seen by the host machine, and one or more mapping values pre-known and associated with the first storage system.

29. The method of claim 25 wherein in step (c) the processing is achieved via algorithm using one or more of the mapping values as variables.

30. The method of claim 25 wherein in step (d) the record is forwarded over a TCP/IP-enabled connection.

31. The method of claim 25 wherein in step (e) the first storage system is accessed for read through a Host Bus Adaptor and Fibre Channel network.

32. The method of claim 25 wherein in step (e) the first storage system is accessed for read through an Ethernet™

network.

33. A method for providing a time-ordered snapshot view of all or a specified portion of a data volume backed up from a first storage system to a second storage system on behalf of a host machine comprising steps of:

(a) at the host machine, ordering a view of the host volume specifying a date and time for the ordered view;

(b) receiving the order at a server adapted to assemble and enable access to the ordered view;

(c) at the server, accessing the second storage system according to the date and time of the ordered view and accessing any specified data writes representing data changes to the volume since a last time interval data writes were stored for the volume;

(d) accessing volume data for that time specific to the view ordered by the host including and associating the change data thereto replacing the writes affected by the change data to assemble a snapshot appropriate to the date and time specified in the order; and

(e) enabling access to the view from the host machine.

34. The method of claim 33 wherein in step (a) the host machine is one of a plurality of host machines that are LAN-connected and the first storage system is a SAN-based storage system.

35. The method of claim 34 wherein in step (a) more than one host machine orders a snapshot view of a same time period.

36. The method of claim 33 wherein in step (b) the order is received over a TCP/IP enabled network connection and the server is the same server that backed up the data to the second storage system.

37. The method of claim 33 wherein in step (c) the second storage system is one of a variant of a random access disk including a SATA or a PATA disk drive.

38. The method of claim 33 wherein in step (d) the change data is stored separately from the volume data and the volume data is rolled forward or backward according to existing time references adopting the change stored for those time references to complete the snapshot view.

39. The method of claim 38 wherein in step (d) a complete volume is rolled forward or backward to complete a full volume view.

40. The method of claim 39 wherein in step (d) a portion of the volume representing application data or specific files is rolled forward or backward to complete an application level or file specific view.

41. The method of claim 33 wherein in step (e) access is enabled by serving the view and access includes mounting the volume for data recovery.

42. The method of claim 33 wherein in step (e) the view is made compatible with the current logical view of the first storage system at the host machine including file system, application system, and platform.

43. A system for providing secondary data storage and recovery services for one or more networked host nodes comprising:

a server application for facilitating data backup and recovery services;

a first data storage medium for primary data backup;

a second data storage medium for secondary data storage of the backup data;

at least one client application for duplicating data write activity by a host to the primary storage medium to a transactional log volume; and

at least one machine instruction enabling direct data read and data clear operations to be performed on the transactional log volume by the server application to backup write data from the log volume to the secondary storage medium.

44. The system of claim 43 wherein the networked host nodes are connected to a local area network (LAN).

45. The system of claim 43 wherein the host nodes include but are not limited to server nodes.

46. The system of claim 43 implemented in conjunction with a storage area network (SAN) accessible through a Fibre Channel network.

47. The system of claim 43 implemented in conjunction with a SAN accessible through an Ethernet™

network.

48. The system of claim 43 wherein the second storage medium is one of a variant of a random access disk including a serial-advanced-technology-attachment (SATA) or a parallel-advanced-technology-attachment (PATA) drive system.

49. The system of claim 43 wherein the transactional log volume is hosted at a system providing the first data storage medium and wherein the log is implemented as a logical unit number device.

50. The system of claim 43 wherein the transactional log volume is hosted at a system providing the second data storage medium and wherein the log is implemented as a virtual logical unit number device in flash memory.

51. The system of claim 43 wherein a bit flagging system is used to enable data write, data read, and data clear operations performed on the transactional log volume.

52. The system of claim 43 wherein the transactional log volume has a logical unit number identifier and is shared by a plurality of host nodes, each of the nodes assigned a sub logical unit number identifier within the log volume.

53. The system of claim 43 wherein there is more than one host, each host having a separate transactional log volume with a logical unit number identifier.