Networked raid in a virtualized cluster

• US 8,001,352 B1
• Filed: 04/16/2008
• Issued: 08/16/2011
• Est. Priority Date: 04/17/2007
• Status: Active Grant

• Current Assignee: AmZetta Technologies, LLC (American Megatrends International, LLC)
• Original Assignee: American Megatrends Incorporated
• Inventors: Grandhi, Suresh, Chatterjee, Paresh, Narayanan, Ajit, Mahalingam, Anandh
• Primary Examiner(s): Farrokh; Hashem
• Application Number:
  US12/104,123
• Time in Patent Office: 1,217 Days
• Field of Search: 711/165, 711/114
• US Class Current: 711/165
• CPC Class Codes: G06F 11/1464 for networked environments
• CPC Class Codes: G06F 12/02 Addressing or allocation; Relocation program address sequencing G06F9/00; arrangements for selecting an address in a digital store G11C8/00
• CPC Class Codes: G06F 3/0607 by facilitating the process of upgrading existing storage systems, e.g. for improving compatibility between host and storage device
• CPC Class Codes: G06F 3/0617 in relation to availability
• CPC Class Codes: G06F 3/0647 Migration mechanisms
• CPC Class Codes: G06F 3/0649 Lifecycle management
• CPC Class Codes: G06F 3/0658 Controller construction arrangements
• 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

Abstract

Technologies are described herein for providing networked RAID in a virtualized storage cluster. The storage capacity of a storage cluster having two or more storage nodes is organized into tiers. A portion of the available storage capacity is allocated to one tier that is organized using chained declustering. Another portion of the available storage capacity is allocated to another tier that is organized using a networked RAID configuration. The storage cluster monitors the frequency at which data in the storage cluster is accessed. Frequently used data that is stored in the network RAID-configured tier is promoted to the chained declustered-configured tier. Infrequently used data that is stored in the chained declustered-configured tier is demoted to the network RAID-configured tier. The zone size for zones stored in the chained declustered-configured tier may be different than for zones stored in the network RAID-configured tier.

Claims

1. A method for utilizing a networked redundant array of inexpensive disks (RAID) configuration in a networked storage cluster having two or more nodes that exposes a storage capacity to one or more initiators, the method comprising:

allocating a first portion of the storage capacity of the storage cluster to a first tier, wherein data stored in the first tier is organized in a chained declustered configuration;

allocating a second portion of the storage capacity of the storage cluster to a second tier, wherein data stored in the second tier is organized in a networked RAID-5 configuration;

monitoring the frequency at which the data stored in the first tier and the second tier is accessed; and

promoting data that is accessed frequently from the second tier to the first tier, and demoting data that is used infrequently from the first tier to the second tier, wherein the first tier is subdivided into zones having a first size, wherein the second tier is subdivided into zones having a second size, and wherein the first size is different than the second size.

2. The method of claim 1, wherein the first tier and the second tier are subdivided into zones, and wherein promoting data that is accessed frequently comprises moving a zone from the second tier to the first tier.

3. The method of claim 1, wherein demoting data that is accessed infrequently comprises moving a zone from the first tier to the second tier.

4. The method of claim 1, wherein the storage cluster comprises N storage nodes, and wherein demoting data that is accessed infrequently comprises:

subdividing a zone stored in the first tier into N−

1 sub-zones;

calculating the parity of the sub-zones; and

storing each of the sub-zones and the calculated parity in the second tier on different nodes of the storage cluster.

5. The method of claim 1, wherein the storage cluster comprises N storage nodes, and wherein promoting data that is accessed frequently comprises:

reading a sub-zone from the second tier of N−

1 of the nodes;

assembling the sub-zones into a zone; and

storing the zone on the first tier of the storage cluster.

6. The method of claim 1, wherein the storage cluster comprises N storage nodes, and wherein demoting data that is accessed infrequently comprises:

subdividing a zone stored in the first tier into N−

1 sub-zones;

subdividing each of the sub-zones into N−

1 stripes;

generating a parity stripe for each sub-zone; and

storing a stripe from each of the sub-zones and a parity stripe from each sub-zone on different nodes in the second tier of the storage cluster.

7. The method of claim 1, wherein the storage cluster comprises N storage nodes, and wherein promoting data that is accessed frequently comprises:

reading N−

1 stripes from each node in the second tier of the storage cluster;

assembling the N−

1 stripes into N−

1 sub-zones;

assembling the sub-zones into a zone; and

storing the assembled zone on the first tier of the storage cluster.

8. The method of claim 1, wherein the data is organized on the first and second tiers such that when a new storage node is added to the storage cluster and data is re-striped between the storage nodes, only data that will subsequently reside in the new storage node is moved to the new storage node during re-striping.

9. The method of claim 8, wherein the data is further organized on the first and second tiers such that when a new storage node is added to the storage cluster and data is re-striped between the storage nodes, no movement of data occurs between two storage nodes that existed in the storage cluster prior to the addition of the new storage node.

10. A computer storage medium having computer-executable instructions stored thereon which, when executed by a computer system, cause the computer system to:

allocate a first portion of a storage capacity to a first tier, the first tier being organized in a chained declustered configuration;

allocate a second portion of a storage capacity to a second tier, the second tier being organized in a networked RAID configuration; and

to

move data that is accessed frequently from the second tier to the first tier, and to move data that is accessed infrequently from the first tier to the second tier, wherein data stored in the first tier is organized into zones having a first size and data stored in the second tier is organized into zones having a second size different from the first size.

11. The computer storage medium of claim 10, wherein moving data that is accessed frequently from the second tier to the first tier comprises moving a zone from the second tier to the first tier, and wherein moving data that is accessed infrequently from the first tier to the second tier comprises moving a zone from the first tier to the second tier.

12. The computer storage medium of claim 11, wherein the computer system comprises N storage nodes, and wherein moving data that is accessed infrequently comprises:

subdividing a zone stored in the first tier into N−

1 sub-zones;

performing an exclusive-or (XOR) operation on the sub-zones; and

storing each of the sub-zones and the results of the XOR operation in the second tier on different storage nodes.

13. The computer storage medium of claim 12, wherein the computer system comprises N storage nodes, and wherein moving data that is accessed frequently comprises:

reading a sub-zone from the second tier of N−

1 of the nodes;

assembling the sub-zones into a zone; and

storing the assembled zone on the first tier of the storage cluster.

14. The computer storage medium of claim 11, wherein the computer system comprises N storage nodes, and wherein moving data that is accessed infrequently comprises:

subdividing a zone stored in the first tier into N−

1 sub-zones;

subdividing each of the sub-zones into N−

1 stripes;

generating a parity stripe for each sub-zone by performing an exclusive-or (XOR) operation on the stripes of each sub-zone; and

storing a stripe from each of the sub-zones and a parity stripe from each sub-zone in the second tier on a different node of the storage cluster.

15. The computer storage medium of claim 14, wherein the computer system comprises N storage nodes, and wherein moving data that is accessed frequently comprises:

reading N−

1 stripes from each node in the second tier of the storage cluster;

assembling the N−

1 stripes into N−

1 sub-zones; and

storing the zone on the first tier of the storage cluster.

16. The computer storage medium of claim 11, wherein the data is organized on the first tier such that when a new storage node is added to the storage cluster and data is re-striped between the storage nodes, only data that will subsequently reside in the new storage node is moved to the new storage node during re-striping.

17. The computer storage medium of claim 16, wherein the data is further organized on the first tier such that when a new storage node is added to the storage cluster and data is re-striped between the storage nodes, no movement of data occurs between two storage nodes that existed in the storage cluster prior to the addition of the new storage node.

18. A method for utilizing a networked redundant array of inexpensive disks (RAID) configuration in a storage cluster having two or more nodes that exposes a storage capacity to one or more initiators, the method comprising:

allocating a first portion of the storage capacity of the storage cluster to a first tier, wherein data stored in the first tier is organized into one or more zones having a first size that are stored in a chained declustered configuration;

allocating a second portion of the storage capacity of the storage cluster to a second tier, wherein data stored in the second tier is organized into one or more zones having a second size different than the first size that are stored in a networked RAID-5 configuration;

monitoring the frequency at which the data stored in the first tier and the second tier is accessed; and

promoting data in a first zone that is accessed frequently by moving the first zone from the second tier to the first tier, and demoting data in a second zone that is used infrequently by moving the second zone from the first tier to the second tier, wherein the storage cluster comprises N storage nodes, and wherein demoting data that is accessed infrequently comprises subdividing a zone stored in the first tier into N−

1 sub-zones, calculating the parity of the sub-zones, and storing each of the sub-zones and the parity on different nodes of the storage cluster.

19. The method of claim 18, wherein promoting data that is accessed frequently comprises reading a sub-zone from the second tier of N−

1 of the nodes, assembling the sub-zones into a zone, and storing the assembled zone on the first tier of the storage cluster.