Virtualization of metadata for file optimization
View Patent ↗Mechanisms are provided for optimizing files while allowing application servers access to metadata associated with preoptimized versions of the files. During file optimization involving compression and/or compaction, file metadata changes. In order to allow file optimization in a manner transparent to application servers, the metadata associated with preoptimized versions of the files is maintained in a metadata database as well as in an optimized version of the files themselves.
1. A method comprising;
receiving a read directory request, the read directory request corresponding to a directory associated with a plurality of files, wherein the plurality of files are optimized using mechanisms including deduplication;
populating a client side cache with metadata fetched from a metadata database for the plurality of files in the directory upon receiving the read directory request;
receiving a subsequent read directory request;
determining whether metadata needed to respond to the subsequent read directory request is available in the client side cache;
accessing a metadata database if the metadata is not available in the client side cache;
comparing timestamps to determine whether an entry in the metadata database is valid, wherein if the entry is not valid, the subsequent read directory request is handled by accessing metadata included in the plurality of files that are optimized using mechanisms including deduplication and maintained using a suitcase file.
2. The method of claim 1 , wherein the metadata database includes file identifiers, access times, and metadata.
3. The method of claim 2 , wherein file identifiers include user identifiers, group identifiers, and inode numbers.
4. The method of claim 2 , wherein access times include times corresponding files were last accessed.
5. The method of claim 2 , wherein metadata includes file names and file sizes associated with preoptimized versions of the plurality of files.
6. The method of claim 1 , wherein the metadata database resides at the application server.
7. The method of claim 1 , wherein the plurality of optimized files are optimized files that are also compressed.
8. The method of claim 1 , wherein the metadata database need not be synchronized with a directory hierarchy.
9. The method of claim 1 , wherein accessing metadata included in one of the plurality of optimized files comprises accessing the suitcase file.
10. The method of claim 9 , wherein the suitcase file includes metadata and compressed data for the plurality of optimized files.
11. A system comprising;
an interface operable to receive a read directory request, the read directory request corresponding to a directory associated with a plurality files, wherein the plurality of files are optimized using mechanisms including deduplication, wherein a client side is populated with metadata fetched from a metadata database for the plurality of files in the directory upon receiving the read directory request;
a processor operable to determine whether metadata needed to respond to the read directory request is available in the client side cache upon receiving a subsequent read directory request, the processor operable to access the metadata database if the metadata is not available in the client side cache;
wherein timestamps are compared to determine whether an entry in the metadata database is valid, wherein if the entry is not valid, the subsequent read directory request is handled by accessing metadata included in the plurality of files that are optimized using mechanisms including deduplication and maintained using a suitcase file.
12. The system of claim 11 , wherein the in the metadata database includes file identifiers, access times, and metadata.
13. The system of claim 12 , wherein file identifiers include user identifiers, group identifiers, and inode numbers.
14. The system of claim 12 , wherein access times include times corresponding files were last accessed.
15. The system of claim 12 , wherein metadata includes file names and file sizes associated with preoptimized versions of the plurality of files.
16. The system of claim 11 , wherein the metadata database resides at the application server.
17. The system of claim 11 , wherein the plurality of optimized files are optimized files that are also compressed.
18. The system of claim 11 , wherein the metadata database need not be synchronized with a directory hierarchy.
19. The system of claim 11 , wherein accessing metadata included in one of the plurality of optimized files comprises accessing the suitcase file.
20. The system of claim 19 , wherein the suitcase file includes metadata and compressed data for the plurality of optimized files.
21. A non-transitory computer readable medium comprising;
computer code for receiving a read directory request, the read directory request corresponding to a directory associated with a plurality of files, wherein the plurality of files are optimized using mechanisms including deduplication;
computer code for populating a client side cache with metadata fetched from a metadata database for the plurality of files in the directory upon receiving the read directory request;
computer code for receiving a subsequent read directory request;
computer code for determining whether metadata needed to respond to the subsequent read directory request is available in the client side cache;
computer code for accessing a metadata database if the metadata is not available in the client side cache;
computer code for comparing timestamps to determine whether an entry in the metadata database is valid, wherein if the entry is not valid, the subsequent read directory request is handled by accessing metadata included in the plurality of files that are optimized using mechanisms including deduplication and maintained using a suitcase file.