Data distribution through capacity leveling in a striped file system

• US 8,117,388 B2
• Filed: 04/30/2009
• Issued: 02/14/2012
• Est. Priority Date: 04/30/2009
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Jernigan, Richard P. IV
• Primary Examiner(s): Dinh, Son
• Application Number:
  US12/433,089
• Publication Number:
  US 20100281214A1
• Time in Patent Office: 1,020 Days
• Field of Search: 711/114, 711/118, 707/10
• US Class Current: 711/114
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/0631 by allocating resources to storage systems
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

A data distribution technique is configured to provide capacity leveling in a striped file system. When a new node is added to a striped volume set, the striping table is evolved to accommodate the newly added node. Each node of a cluster is illustratively associated with a capacity value that takes into account, e.g., processor speed, number of processors, hardware configuration and/or software available for the node. During the evolution process of the striping table, the technique apportions stripes of the SVS among the nodes in a manner so that they are optimally assigned to the nodes in accordance with each node'"'"'s capacity value. By utilizing the evolutionary striping table that incorporates capacity values, heterogeneous nodes may be utilized to their maximum capacity within a striped volume set, thereby reducing underutilized processing resources.

Claims

1. A method for capacity leveling in a clustered computer system supporting a striped file system comprising:

assigning a capacity value associated with each node of a plurality of nodes organized as the clustered computer system to support the striped file system;

iteratively adjusting, by a volume striping module executing on a processor of one of the plurality of nodes, a number of occurrences of a node within a striping data structure to generate a capacity leveled striping data structure based on the capacity value associated with each node; and

utilizing the capacity leveled striping data structure to perform data allocation in the striped file system in response to data access requests.

2. The method of claim 1 wherein each node appears in the striping data structure in approximate proportion to its capacity value divided by a total of all capacity values of the plurality of nodes.

3. The method of claim 1 further comprising utilizing a semi-static distribution technique to iteratively adjust the number of occurrences of the node within the striping data structure.

4. The method of claim 1 further comprising performing a restriping operation on the striped file system to utilize the capacity leveled striping data structure.

5. The method of claim 1 further comprising determining each node'"'"'s capacity value using a processor speed of the node.

6. The method of claim 1 wherein the plurality of nodes are associated with differing capacity values.

7. The method of claim 1 further comprising implementing a striped volume set of the striped file system wherein the striped volume set comprises a plurality of volumes with each volume being stored on one of the plurality of nodes.

8. The method of claim 1 further comprising storing an evolved striping data structure in a volume location database that is accessible from each of the plurality of nodes.

9. The method of claim 1 wherein data allocation comprises distributing data across the plurality of nodes in accordance with the capacity leveled striping data structure.

10. A method for generating a capacity leveled striping data structure for use in a clustered computer system supporting a striped file system comprising:

assigning a capacity value associated with each node of a plurality of nodes organized as the clustered computer system to support the striped file system, wherein the capacity value reflects a determination of processing speed associated with a node as compared to other nodes of the plurality of nodes;

iteratively adjusting, by a volume striping module executing on a processor of one of the plurality of nodes, a number of occurrences of a node within a striping data structure to generate the capacity leveled striping data structure based on the capacity value associated with each node wherein the capacity leveled striping data structure contains each node in proportion to the node'"'"'s capacity value divided by a sum of capacity values of all nodes in the clustered computer system; and

utilizing, by the volume striping module, the capacity leveled striping data structure to allocate data among the nodes of the clustered computer system.

11. A system comprising:

a set of nodes, each comprising a processor, organized as a cluster configured to support a striped file system comprising a striped volume set apportioned among volumes stored on the set of nodes, wherein each of the set of nodes is associated with a capacity value; and

a volume striping module, executing on one of the processors, configured to apportion stripes of the striped volume set among the volumes stored on the set of nodes so that each node appears in a striping data structure in approximate proportion to the node'"'"'s capacity value divided by a total of all capacity values of the set of nodes, wherein the striped file system is further configured to utilize the striping data structure to perform data allocation on the striped file system in response to data access requests.

12. The system of claim 11 wherein the striping data structure is stored in a volume location database accessible by each of the set of nodes.

13. The system of claim 11 wherein each node of the set of nodes is assigned a differing capacity value.

14. The system of claim 11 wherein the volume striping module is further configured to evolve the striping data structure in response to addition of a new node to the set of nodes, wherein the striping data structure comprises entries for each of the set of nodes and the new node in approximate proportion to each node'"'"'s capacity value divided by a total of all capacity values of the set of nodes and the new node.

15. The system of claim 14 wherein the volume striping module is further configured to perform a restriping operation on the striped volume set using the striping data structure.

16. The system of claim 14 wherein the volume striping module is further configured to evolve the striping data structure by iteratively adjusting a number of occurrences of a node within the striping structure using a semi-static distribution technique.

17. A method comprising:

adding a new node to a set of nodes, each comprising at least a processor, configured as a cluster to support a striped file system of a storage system;

evolving a striping data structure associated with the striped file system to incorporate the new node, wherein the evolved striping data structure comprises entries for each of the set of nodes and the new node in approximate proportion to a capacity value associated with each of the nodes divided by a total of all capacity values of the set of nodes and the new node; and

utilizing the capacity leveled striping data structure to perform data allocation on the striped file system in response to data access requests.

18. The method of claim 17 further comprising performing a restriping operation on the striped file system using the evolved striping data structure.

19. The method of claim 17 further comprising storing the striping data structure in a volume location database accessible by each node of the set of nodes and the new node.

20. The method of claim 17 wherein the evolving of the striping data structure utilizes a semi-static distribution technique.

21. The method of claim 17 wherein evolving a striping data structure comprises identifying a capacity value for the new node to be added and dividing the identified capacity value by a total capacity value of the new node to be added and all additional nodes to be added to the striping data structure.