METHOD AND SYSTEM FOR HETEROGENEOUS DATA VOLUME

• US 20120017043A1
• Filed: 09/21/2011
• Published: 01/19/2012
• Est. Priority Date: 07/07/2010
• Status: Active Grant

• Current Assignee: Nexenta by DDN, Inc.
• Original Assignee: Nexenta Systems Incorporated (DataDirect Networks Incorporated)
• Inventors: AIZMAN, Alexander, REMANAN PILLAI, Vineeth
• Granted Patent: US 8,954,669 B2
• Time in Patent Office: Days
• US Class Current: 711/114
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/3048 where the topology of the computing system or computing system component explicitly influences the monitoring activity, e.g. serial, hierarchical systems
• CPC Class Codes: G06F 2211/1028 Distributed, i.e. distributed RAID systems with parity
• CPC Class Codes: G06F 2211/103 Hybrid, i.e. RAID systems with parity comprising a mix of RAID types
• CPC Class Codes: G06F 3/061 Improving I/O performance
• 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 3/0673 Single storage device
• CPC Class Codes: G06F 3/0685 Hybrid storage combining heterogeneous device types, e.g. hierarchical storage, hybrid arrays
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD
• CPC Class Codes: G06F 7/00 Methods or arrangements for processing data by operating upon the order or content of the data handled logic circuits H03K19/00

Abstract

A method and system is disclosed for providing a heterogeneous data storage comprising a plurality of storage devices of different types with respect to device vendor, model, capacity, performance and/or function. The present invention employs data access mechanisms specific to the type of underlying storage and the type of data to be stored or retrieved, and provides for integrated remote mirroring, disaster recovery and hierarchical storage management (HSM), as well as improved I/O performance and life expectancy of storage disks. A method of writing to and reading from heterogeneous data volume is also disclosed.

Claims

1. A system comprising:

a processor;

a heterogeneous data volume comprising a plurality of redundancy groups and storage disks, wherein the storage disks are each associated with a class of storage that reflects the storage disk'"'"'s characteristics, capabilities, or the type of disk attachment; and

an electronic storage medium containing program instructions executable on the processor for writing data to and reading data from the heterogeneous data volume.

2. The system of claim 1, wherein the disk'"'"'s characteristics and capabilities include any of, or any combination of:

disk vendor, model, capacity, I/O latency or I/O throughput.

3. The system of claim 1, wherein the type of disk attachment may be local or remote.

4. The system of claim 1, wherein the class of storage is associated with specific data access mechanisms used to read data from and write data to the corresponding storage disks of the heterogeneous data volume.

5. The system of claim 1, wherein each of the plurality of redundancy groups is associated with a class of storage, wherein each of the plurality of redundancy groups is comprised of a plurality of other redundancy groups or storage disks.

6. The system of claim 5, wherein the class of storage of a redundancy group is derived from the classes of storage of contained storage disks and contained redundancy groups.

7. The system of claim 1, wherein classes of storage associated with storage disks and redundancy groups are assigned by system administrator, an automated storage management system (SMS), or by the system itself.

8. The system of claim 1, wherein the plurality of redundancy groups are classified into homogenous redundancy groups and heterogeneous redundancy groups.

9. The system of claim 8, wherein each of the heterogeneous redundancy groups comprises storage disks or other redundancy groups that have different classes of storage.

10. The system of claim 8, wherein each of the homogeneous redundancy groups comprises a plurality of identical disks, or a plurality of other identical redundancy groups.

11. A method for writing a logical block to a heterogeneous data volume, the method comprising:

receiving a write operation to write the logical block to the heterogeneous data volume;

determining redundancy groups to be used for storing the logical block in accordance with the class of storage of the logical block and associated number of copies of the logical block to be stored; and

executing the write operation on the selected redundancy groups.

12. The method of claim 11, further comprising:

segmenting the logical block into data and parity blocks in accordance with replication schemes of the redundancy groups;

selecting data writing mechanisms in accordance with classes of storage of the storage disks in the redundancy groups; and

writing the data and parity blocks to the storage disks using the data write mechanism for the class of storage of the disk being written.

13. The method of claim 11 that further comprises determining whether metadata associated with the logical block needs to be stored and writing the metadata associated with the logical block, if necessary.

14. The method of claim 11, wherein the class of storage for the logical block is assigned by one of a system, SMS, or by the application that initiated the write operation.

15. The method of claim 13, wherein the metadata associated with the logical block and stored on the heterogeneous data volume comprises the class of storage of the logical block and location of extra copies of the logical block.

16. The method of claim 12, wherein the data writing mechanism comprises selecting a logical block for matching the class of storage of the logical block with classes of storage of those respective redundancy groups and storage disks.

17. A method of reading data from a heterogeneous data volume, the method comprising:

receiving a read operation to read a logical block from the heterogeneous data volume;

determining redundancy groups that contain the logical block; and

executing the read operation on the determined redundancy groups.

18. The method of claim 17, further comprising:

determining whether the logical block has to be segmented and read in parallel from the redundancy groups that contain the logical block;

determining the number of data segments to be used for reading the logical block;

determining read preference values of selected redundancy groups;

selecting a redundancy group or storage disk to read the logical block;

determining data reading mechanisms associated with each class of storage of the storage disks;

reading data segments using the data reading mechanisms; and

combining the read data segments to form the logical block.

19. The method of claim 18, wherein the associated data reading mechanism is used to read the data segment from each selected storage disk.

20. The method of claim 17, wherein determining of redundancy groups includes matching of the class of storage of the logical block with the classes of storage of the redundancy groups.

21. The method of claim 20, wherein determining of redundancy groups may further include processing of the metadata associated with the logical block.

22. The method of claim 17, wherein the class of storage of the logical block is assigned by a system, SMS, or by the application that initiated the read operation.

23. The method of claim 17, wherein the class of storage of the logical block is retrieved from the metadata associated with the logical block.

24. The method of claim 18, wherein multiple copies of the logical block stored on the heterogeneous data volume are used to read the logical block in segments in parallel from the corresponding redundancy groups and storage disks that hold the copies.

25. The method of claim 18, wherein the read preference value is computed as a function of runtime read latency or number of pending I/O operations on the corresponding redundancy group or storage disk.

26. The method of claim 18, wherein the read preference value is assigned by the system administrator, SMS, or is a property of the class of storage associated with the respective redundancy group and storage disk.