File system metadata deduplication

1. A method, comprising:

analyzing file metadata structures of a file system, wherein each of the file metadata structures includes a plurality of metadata elements;

identifying a metadata element of the plurality of metadata elements that is duplicated among the analyzed file metadata structures, wherein instances of the identified metadata element reference a same data chunk, wherein the instances of the identified metadata element at least include a first leaf node of a first file metadata structure of the file metadata structures and a second leaf node of a second file metadata structure of the file metadata structures; and

deduplicating the instances of the identified metadata element at least in part by modifying the second file metadata structure to reference the first leaf node of the first file metadata structure of the file metadata structures instead of referencing the same data chunk,

wherein modifying the second file metadata structure to reference the first leaf node of the first file metadata structure includes modifying a parent metadata element of the second leaf node of the second file metadata structure to reference the first leaf node of the first file metadata structure instead of referencing the second leaf node of the second file metadata structure and deleting the second leaf node of the second file metadata structure, and

wherein the first leaf node of the first file metadata structure is an instance of the identified metadata element among the instances of the identified metadata element having a highest reference count.

2. The method of claim 1 , wherein analyzing file metadata structures of the file system comprises scanning a bottom level of the file metadata structures and a level above the bottom level of the file metadata structures.

3. The method of claim 1 , wherein at least one of the first file metadata structure and the second file metadata structure corresponds to a file generated by a storage cluster.

4. The method of claim 1 , wherein at least one of first file metadata structure and the second file metadata structure corresponds to a file backed up from a primary system to a storage cluster.

5. The method of claim 1 , further comprising identifying a plurality of file metadata structures from a set of stored file metadata structures that include the metadata element.

6. The method of claim 1 , wherein the metadata element is configured to store a value, wherein at least two of the analyzed file metadata structures include the metadata element that stores the value.

7. The method of claim 1 , wherein deduplicating instances of the identified metadata element includes identifying which instance of the identified metadata element has the highest reference count from a plurality of instances of the identified metadata element.

8. The method of claim 7 , wherein a reference count indicates a number of one or more other metadata elements that reference the metadata element.

9. The method of claim 1 , wherein deduplicating instances of the identified metadata element comprises deleting one or more instances of the identified metadata element not having the highest reference count.

10. The method of claim 9 , wherein the one or more instances of the identified metadata element not having the highest reference count are deleted from a solid state disk of a storage cluster.

11. The method of claim 1 , wherein the identified metadata element is deduplicated as a background process of a storage cluster.

12. The method of claim 11 , wherein the storage cluster is comprised of a plurality of storage nodes, wherein the metadata associated with a plurality of files is stored across the plurality of storage nodes.

13. The method of claim 1 , wherein deduplicating instances of the identified metadata element comprises determining a corresponding reference count associated with each instance of the identified metadata element.

14. The method of claim 13 , wherein the determined reference count is the same for each instance of the identified metadata element.

15. The method of claim 14 , wherein modifying the second file metadata structure includes selecting to reference the first leaf node of the first file metadata structure.

16. A computer program product, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

analyzing file metadata structures of a file system, wherein each of the file metadata structures includes a plurality of metadata elements;

identifying a metadata element of the plurality of metadata elements that is duplicated among the analyzed file metadata structures, wherein instances of the identified metadata element reference a same data chunk, wherein the instances of the identified metadata element at least include a first leaf node of a first file metadata structure of the file metadata structures and a second leaf node of a second file metadata structure of the file metadata structures; and

deduplicating the instances of the identified metadata element at least in part by modifying the second file metadata to reference the first leaf node of the first file metadata structure of the file metadata structures instead of referencing the same data chunk,

wherein modifying the second file metadata structure to reference the first leaf node of the first file metadata structure includes modifying a parent metadata element of the second leaf node of the second file metadata structure to reference the first leaf node of the first file metadata structure instead of referencing the second leaf node of the second file metadata structure and deleting the second leaf node of the second file metadata structure, and

wherein the first leaf node of the first file metadata structure is an instance of the identified metadata element among the instances of the identified metadata element having a highest reference count.

17. A system, comprising:

a processor configured to:

analyze file metadata structures of a file system, wherein each of the file metadata structures includes a plurality of metadata elements;

identify a metadata element of the plurality of metadata elements that is duplicated among the analyzed file metadata structures, wherein instances of the identified metadata element reference a same data chunk, wherein the instances of the identified metadata element at least include a first leaf node of a first file metadata structure of the file metadata structures and a second leaf node of a second file metadata structure of the file metadata structures; and

deduplicate the instances of the identified metadata element at least in part by modifying the second file metadata structure to reference the first leaf node of the first file metadata structure of the file metadata structures instead of referencing the same data chunk,

wherein modifying the second file metadata structure to reference the first leaf node of the first file metadata structure includes modifying a parent metadata element of the second leaf node of the second file metadata structure to reference the first leaf node of the first file metadata structure instead of referencing the second leaf node of the second file metadata structure and deleting the second leaf node of the second file metadata structure, and

wherein the first leaf node of the first file metadata structure is an instance of the identified metadata element among the instances of the identified metadata element having a highest reference count; and

a memory coupled to the processor and configured to provide the processor with instructions.