Extended attribute storage

• US 10,642,789 B2
• Filed: 01/15/2015
• Issued: 05/05/2020
• Est. Priority Date: 11/01/2014
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Inventors: Panigrahy, Padmagandha, Sachan, Abhay, Narra, Chaitanya, Venkateshappa, Amarish Shapur, Kamalaksha, Venkataraman, Boogarapu, Anil Kumar, Chellam, Rajagopal
• Primary Examiner(s): Ortiz Ditren, Belix M
• Application Number:
  US15/521,561
• Publication Number:
  US 20170316024A1
• Time in Patent Office: 1,937 Days
• Field of Search: None
• CPC Class Codes: G06F 16/13 File access structures, e.g. distributed indices arrangements of input from, or output to, record carriers G06F3/06
• CPC Class Codes: G06F 16/137 Hash-based content-based indexing of textual data G06F16/31
• CPC Class Codes: G06F 3/0604 Improving or facilitating administration, e.g. storage management
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0673 Single storage device

Abstract

In one example, a request for storage of an extended attribute (EA) of a file system object is received. A storage location for the EA is determined from a plurality of storage locations, based on one or more characteristics of the EA, the plurality of storage locations including an inode, a first storage file accessible by a first B+ tree, and a second storage file accessible by a second B+ tree. The EA is accordingly stored in the determined storage location.

Claims

1. A system comprising:

a processor;

a file system coupled to the processor to receive a request for storage of an extended attribute (EA) of a file system object; and

a non-transitory computer readable medium coupled to the processor and storing instructions that, when executed, cause the processor to;

determine a storage location for the EA from a plurality of storage locations, based on one or more characteristics of the EA and based on at least one of an EA storage map that indicates the storage location based on the characteristics of the EA and a size criteria that is based on a size of the EA and a predefined size of the inode, wherein the plurality of storage locations include an inode, a first storage file accessible by a first B+ tree, and a second storage file accessible by a second B+ tree;

store the EA in the inode, based on the determined storage location;

de-duplicate and store the EA in the first storage file using the first B+ tree, based on the determined storage location; and

store the EA in the second storage file using the second B+ tree, based on the determined storage location.

2. The system as claimed in claim 1, wherein the inode is one of a plurality of secondary inodes co-located with a primary inode, wherein a number of the secondary inodes and a size of each of the secondary inodes are predefined.

3. The system as claimed in claim 1, wherein the characteristics of the EA comprise at least one of a name of the EA, a user defined characteristic indicative of the storage location, and a size of the EA.

4. The system as claimed in claim 1, wherein the instructions, when executed, cause the processor to de-duplicate and store the EA in the first storage file by increasing a reference count of an existing EA in the first storage file when the existing EA has the same hash value and EA value as the EA to be stored.

5. The system as claimed in claim 1, wherein the instructions, when executed, cause the processor to determine the storage location in which the EA is stored based on the one or more characteristics of the EA and determine an EA value corresponding to the EA from the storage location.

14. The system as claimed in claim 1, wherein the instructions that cause the processor to de-duplicate and store the EA in the first storage file using the first B+ tree include instructions to cause the processor to:

determine a hash value of an EA value of the EA to be stored;

determine, from the first B+ tree, if a key node having an identical hash value to that of the EA value exists;

determine whether a corresponding EA value of the key node having the identical hash value is identical to the EA value of the EA to be stored; and

increment a reference count of the corresponding EA value by one when determined to be identical to the EA value of the EA to be stored.

15. The system as claimed in claim 1, wherein the instructions that cause the processor to store the EA in the second storage file using the second B+ tree include instructions to cause the processor to:

determine a hash value of a name of the EA;

determine, from the second B+ tree, if a key node exists with an identical hash value to that of the name of the EA and an identical EA name to that of the EA;

update an EA value corresponding to the key node in a second storage file in response to a determination that the key node exists with identical hash value and identical EA name; and

store the EA, as a key-value pair appended to a header, with a new key node in response to a determination that no key node exists with the identical hash value.

16. The system as claimed in claim 1, wherein the size criteria includes whether the size of the EA is less than the predefined size of the inode and if a total size of the EA and existing EAs in the inode is less than the predefined size of the inode.

6. A method of storing an extended attribute (EA) corresponding to a file system object managed by a file system, the method comprising:

determining if at least one of a plurality of characteristics of the extended attribute is indicative of storage of the extended attribute using a first B+ tree that de-duplicates and stores the extended attribute, and storing the extended attribute using the first B+ tree based on the determination;

determining if at least one of the plurality of characteristics of the extended attribute is indicative of storage of the extended attribute in an inode, and storing the extended attribute in the inode of the file system object based on the determination; and

storing the extended attribute using a second B+ tree if at least one of the plurality of characteristics of the extended attribute does not provide an indication for storage using the first B+ tree or in the inode of the file system object.

7. The method as claimed in claim 6, wherein the storing the extended attribute using the first B+ tree comprises:

determining a hash value of a value of the EA to be stored;

determining, from the first B+ tree, if a key node having an identical hash value exists;

determining whether a corresponding EA value of the key node having the identical hash value is identical to the value of the EA to be stored; and

incrementing a reference count of the corresponding EA value by one when found to be identical to the value of the EA to be stored.

8. The method as claimed in claim 7, further comprising:

incrementing the hash value of the value of the EA by one when the corresponding EA value of the key node having the identical hash value is not identical to the value of the EA to be stored; and

determining, from the first B+ tree, if another key node having an identical hash value to the incremented hash value exists.

9. The method as claimed in claim 7, further comprising:

adding a new key node corresponding to the value of the EA to be stored if a key node having the identical hash value does not exist;

determining, from a first free space B+ tree, an offset location at which the EA is to be stored in a first storage file associated with the first B+ tree; and

storing the EA, as a key-value pair appended to a header, in the first storage file at the offset location.

10. The method as claimed in claim 6, wherein the storing the extended attribute using the second B+ tree comprises:

determining a hash value of a name of the EA;

determining, from the second B+ tree, if a key node having an identical hash value and identical EA name exists;

updating an EA value corresponding to the key node in a second storage file when the identical hash value and the identical EA name exists; and

storing the EA, as a key-value pair appended to a header, with a new key node when the identical hash value does not exist.

11. A non-transitory computer-readable medium comprising instructions for storing an extended attribute (EA) of a file system object managed by a file system, the instructions being executable by a processing resource to:

receive a request for storing the EA;

determine if a characteristic of the EA indicates one of a de-duplicated storage in a first storage file and storage in an inode, wherein the characteristic of the EA includes a name of the EA and a user defined characteristic, and wherein the first storage file is accessible by a first B+ tree;

store the EA in one of the first storage file with de-duplication and the inode based on the determination; and

store the EA in one of a second storage file and the inode based on a size criteria, if the characteristic of the EA does not indicate one of the de-duplicated storage in the first storage file and the storage in the inode, wherein the second storage file is accessible by a second B+ tree.

12. The non-transitory computer-readable medium of claim 11, wherein the size criteria includes determining if a size of the EA is less than a predefined size of the inode and if a total size of the EA and existing EAs in the inode is less than the predefined size of the inode.

13. The non-transitory computer-readable medium of claim 11, wherein the instructions to store the EA in the first storage file with de-duplication comprise instructions to increment a reference count of the EA in the first storage file by one upon identifying that an existing EA having the same hash value and EA value as the EA to be stored exists in the first storage file.