PARALLEL ACCESS VIRTUAL TAPE LIBRARY AND DRIVES

• US 20090063765A1
• Filed: 09/02/2008
• Published: 03/05/2009
• Est. Priority Date: 08/30/2007
• Status: Active Grant

• Current Assignee: CommVault Systems Incorporated
• Original Assignee: CommVault Systems Incorporated
• Inventors: Kottomtharayil, Rajiv, Muller, Marcus S., Retnamma, Manoj K. Vijayan
• Granted Patent: US 8,706,976 B2
• Time in Patent Office: Days
• US Class Current: 711/112
• CPC Class Codes: G06F 3/0604 Improving or facilitating administration, e.g. storage management
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0622 in relation to access
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/0664 at device level, e.g. emulation of a storage device or system
• CPC Class Codes: G06F 3/0686 Libraries, e.g. tape libraries, jukebox

Abstract

A system and method described herein allows a virtual tape library (VTL) to perform multiple simultaneous or parallel read/write or access sessions with disk drives or other storage media, particularly when subject to a sequential SCSI-compliant layer or traditional limitations of VTLs. In one embodiment, a virtualizing or transaction layer can establish multiple sessions with one or more clients to concurrently satisfy the read/write requests of those clients for physical storage resources. A table or other data structure tracks or maps the sessions associated with each client and the location of data on the physical storage devices.

Claims

1. A virtual tape library (VTL) system for storing data, wherein the VTL receives sequential access read commands and sequential access write commands from two or more client computers, and wherein the VTL reads data from or writes data to at least two non-sequential access data storage devices associated with the VTL, the system comprising:

two or more non-sequential access data storage devices;

at least one data storage device driver coupled to communicate with the two or more non-sequential access data storage devices, wherein the at least one data storage device driver is configured to provide read commands and write commands to the two or more non-sequential access data storage devices;

a disk subsystem coupled to communicate with the at least one data storage device driver; and

,

an interface coupled to receive sequential access read commands from the two or more client computers, and to receive sequential access write commands from the two or more client computers,

wherein the interface is configured to pass to the disk subsystem a non-sequential access read command in response to a received sequential access read command from one of the client computers, even if a sequential data storage device identified by the received sequential access read command is currently occupied with a read or write operation, and even if a removable, sequential access, data storage medium identified in the received sequential access read command is currently being read from or written to by the VTL system,

wherein the interface is also configured to pass to the disk subsystem a non-sequential access write command in response to a received sequential access write command from one of the client computers, even if a sequential data storage device identified by the received sequential access write command is currently occupied with a read or write operation, and even if a removable, sequential access, data storage medium identified in the received sequential access write command is currently being read from or written to by the VTL system,

wherein the disk subsystem is further configured to provide to the at least one data storage device driver a read command in response to the non-sequential access read command received from the interface, and wherein the disk subsystem is also configured to provide to the at least one data storage device driver a write command in response to the non-sequential access write command received from the interface, and

wherein the interface is configured to store and manage a data structure, wherein the data structure tracks

received sequential access read commands from the two or more client computers,

received sequential access write commands from the two or more client computers,

non-sequential access read commands passed to the disk subsystem, and

non-sequential access write commands passed to the disk subsystem.

2. The system of claim 1, wherein the interface further comprises:

a client interface configured to receive sequential read commands and sequential write commands from the two or more client computers;

a session mapper configured to store and manage the data structure and, in response to each received sequential read command and each received sequential write command, to provide a session handle, wherein the client interface passes the session handle to one of the client computers;

wherein the data structure comprises one or more data object mapping tables, wherein the mapping tables are configured to track session handles with corresponding non-sequential access read and write commands passed to the disk subsystem, and to track physical locations of data read from and written to the at least two non-sequential access data storage devices associated with the VTL.

3. The system of claim 1, further comprising an intermediate cache to cache the received sequential access read commands and the received sequential access write commands, wherein the sequential access read commands and the sequential access write commands are sequential SCSI commands, wherein the least two non-sequential access data storage devices are magnetic disk drives, wherein the sequential data storage device is a magnetic tape drive, and wherein the removable, sequential access, data storage medium is a magnetic tape cartridge.

4. The system of claim 1, wherein the disk subsystem further comprises at least one media agent to execute at least one subroutine for performing further data manipulation, wherein the further data manipulation includes data encryption, data decryption, data compression, data decompression, error correction, or data de-duplication.

5. A computer-implementable method of processing commands for a virtual tape library (VTL), wherein the VTL receives sequential access read commands and sequential access write commands from two or more host computers, and wherein the VTL reads data from or writes data to at least two random access data storage devices within the VTL, the method comprising:

intercepting sequential access read commands and sequential access write commands from the two or more host computers, wherein the sequential access read commands and the sequential access write commands from the two or more host computers are commands to read data from or write data to sequential data storage media,

wherein each sequential access read command and each sequential access write command includes associated parameters, and wherein the associated parameters specify a particular sequential access data storage drive and a piece of removable, sequential access, data storage media;

in response to each intercepted sequential access read command and each intercepted sequential access write command from the two or more host computers, returning a session identifier to each host computer,

wherein if the same piece of removable, sequential access, data storage media is requested by two different host computers, then a different session handle is provided to each of the two different host computers, and

wherein if the same sequential access data storage drive is requested by two different host computers, then a different session handle is provided to each of the two different host computers;

tracking, via a data structure, the session identifiers, the parameters, the at least two random access data storage devices within the VTL, and each intercepted sequential access read command and each intercepted sequential access write command from the two or more host computers;

in response to each intercepted sequential access read command and each intercepted sequential access write command from the two or more host computers, performing read or write commands with respect to the at least two random access data storage devices within the VTL;

determining whether any of the intercepted sequential access read commands or the intercepted sequential access write commands have been completed; and

,

if any one of the intercepted sequential access read commands or intercepted sequential access write commands has been completed, then releasing a session associated with a completed sequential access read command or a completed sequential access write command.

6. The computer-implementable method of claim 5, further comprising:

in response to intercepted sequential access read commands and intercepted sequential access write commands, assigning one or more threads to manage different read or write requests from the same host computer.

7. The computer-implementable method of claim 5, further comprising:

performing at least one additional data storage operation on data to be written to the at least two random access data storage devices within the VTL, wherein the additional data storage operation is single instancing, classifying data, load balancing, or generating snapshots.

8. The computer-implementable method of claim 5, further comprising:

generating multiple snapshots of the at least two random access data storage devices within the VTL, wherein each of the multiple snapshots is generated at a different time; and

providing a user interface for displaying and permitting user access to data associated with each of the multiple snapshots.

9. The computer-implementable method of claim 5, further comprising:

generating multiple snapshots of the data structure, wherein each of the multiple snapshots is generated at a different time; and

providing a user interface for displaying and permitting user access to data associated with each of the multiple snapshots.

10. A tangible computer-readable medium containing code, which when executed by a computer causes a virtual tape library (VTL) to perform a method to permit parallel access to two or more drives associated with the VTL, the method comprising:

intercepting sequential access read commands and sequential access write commands from two or more host computers, wherein the sequential access read commands and the sequential access write commands from the two or more host computers are commands to read data from or write data to sequential data storage media,

wherein each sequential access read command and each sequential access write command includes associated parameters, and wherein the associated parameters specify a particular sequential access data storage drive and a piece of removable, sequential access, data storage media;

in response to each intercepted sequential access read command and each intercepted sequential access write command from the two or more host computers, returning a session identifier to each host computer,

wherein if the same piece of removable, sequential access, data storage media is requested by two different host computers, then providing a different session handle to each of the two different host computers, and

wherein if the same sequential access data storage drive is requested by two different host computers, then providing a different session handle to each of the two different host computers.

11. The computer-readable medium of claim 10, further comprising:

tracking, via a data structure, the session identifiers, the parameters, at least two random access data storage devices within the VTL, and each intercepted sequential access read command and each intercepted sequential access write command from the two or more host computers;

in response to each intercepted sequential access read command and each intercepted sequential access write command from the two or more host computers, performing read or write commands with respect to the at least two random access data storage devices within the VTL;

determining whether any of the intercepted sequential access read commands or the intercepted sequential access write commands have been completed; and

,

if any one of the intercepted sequential access read commands or intercepted sequential access write commands has been completed, then releasing a session associated with a completed sequential access read command or a completed sequential access write command.

12. In a virtual tape library (VTL) system for storing data, wherein the VTL receives sequential access read commands and sequential access write commands from at least first and second client computers, and wherein the VTL reads data from or writes data to at least one random access data storage device associated with the VTL, a system comprising:

a transaction layer coupled to receive sequential access read commands and sequential access write commands from the at least first and second client computers,

wherein the transaction layer is configured to establish a first read/write session with the first client computer, and a second read/write session with the second client computer,

wherein the first and second read/write sessions are substantially concurrently implemented, and

wherein the first and second read/write sessions satisfy read/write requests for the at least one random access data storage device associated with the VTL.

13. The system of claim 12, wherein the transaction layer employs a data structure to map sessions associated with the first and second client computers and locations of data on the at least one random access data storage device associated with the VTL.

14. A system of processing commands with respect to a virtual tape library (VTL), wherein the VTL receives sequential access read commands and sequential access write commands from two or more host computers, and wherein the VTL reads data from or writes data to at least one random access data storage device within the VTL, the system comprising:

means for receiving sequential access read commands and sequential access write commands, wherein the sequential access read commands and the sequential access write commands are commands to read data from or write data to sequential data storage media,

wherein each sequential access read command and each sequential access write command includes associated parameters, and wherein the associated parameters specify a particular sequential access data storage drive and a piece of removable, sequential access, data storage media;

means for performing read commands and write commands with respect to the at least one random access data storage device within the VTL, wherein the means for performing read command and write commands is configured

to permit substantially concurrent reads of the at least one random access data storage device within the VTL,

to permit substantially concurrent writes to the at least one random access data storage device within the VTL, and

to permit substantially concurrent reads of and writes to the at least one random access data storage device within the VTL.

15. The system of claim 14, further comprising:

means for providing a session identifier to each host computer,

wherein if the same piece of removable, sequential access, data storage media is requested by two different host computers, then a different session handle is provided to each of the two different host computers, and

wherein if the same sequential access, data storage drive is requested by two different host computers, then a different session handle is provided to each of the two different host computers;

means for tracking, via a data structure, the session identifiers, the parameters, the at least one random access data storage device within the VTL, and each sequential access read command and each sequential access write command from the two or more host computers.

16. The system of claim 14, further comprising:

client-side application means for managing write commands to and read commands from the at least one random access data storage device within the VTL.

17. The system of claim 14, wherein the means for performing read commands and write commands is further configured to permit substantially concurrent reads of or writes to the same piece of removable, sequential access, data storage media.

18. The system of claim 14, wherein the means for performing read commands and write commands is further configured to provide an application program interface for a file system associated with the VTL.

19. The system of claim 14, wherein the system is a data storage appliance, and the means for receiving and the means for performing are housed together in an enclosure with the at least one random access data storage device.

20. A data storage apparatus, wherein the data storage apparatus is coupled to a network and receives sequential access read commands and sequential access write commands from at least first and second client computers, the apparatus comprising:

at least one random access data storage device;

a file system for tracking locations of data on the at least one random access data storage device, wherein the file system is configured to receive the sequential access read commands and sequential access write commands from the at least first and second client computers;

a data storage subsystem configured to receive commands from the file system and in response thereto, to read data from or write data to the at least one random access data storage device; and

,

a single enclosure for carrying the at least one random access data storage device, the file system for tracking locations of data on the at least one random access data storage device, and the data storage subsystem;

wherein the file system is configured to establish a first read/write session with the first client computer, and a second read/write session with the second client computer, and

wherein the least one random access data storage device and the data storage subsystem are configured to perform substantially concurrent read/write operations in response to the established first and second read/write sessions.

21. The apparatus of claim 20, wherein the data storage subsystem data is further configured to perform additional data manipulation operations, wherein the additional data manipulation operations comprise:

data encryption, data decryption, data compression, data decompression, error correction, data de-duplication, single instancing, classifying data, load balancing, or generating snapshots.

22. A method or accessing data on a virtual tape library, the method comprising:

obtaining and storing multiple snapshots of the virtual tape library, wherein each of the multiple snapshots represent virtual tape, and wherein the multiple snapshots are taken sequentially over a period of time and are associated with a point in time at which a snapshot was taken;

maintaining a list of the multiple snapshots;

presenting a user interface with the list of the multiple snapshots;

receiving a user selection of one of the multiple snapshots associated with a selected point in time;

identify data blocks associated with the selected snapshot from the virtual tape library; and

,

presenting, to the user, data from the virtual tape library, wherein the data reflects a state of the virtual tape library at the selected point in time.

23. The method of claim 22 wherein the list tracks individual files associated with the multiple snapshots, wherein the user selection identifies an earlier version of a file associated with the selected snapshot, and wherein presenting to the user data includes presenting the earlier version of the file based on the selected snapshot.