Common template file system tree for virtual environments and virtual servers
A common template file system tree is utilized by isolated operating system processes groups for effective read-only common file set access via multiple file system paths. Files are opened from different views of the file system template for write-ability access and copied into a private modification area; this is also convenient for subsequent online changes and replication.
1. A computer data storage file system comprising:
a storage device;
a namespace;
a plurality of groups of isolated operating system processes;
a common template file system tree stored on the storage device and concurrently accessed by the groups of isolated of operating system processes of the groups of isolated operating system processes using a view to the common template file system tree;
a first set of names from the namespace for direct access of contents of the common template file system tree by the groups of isolated operating system processes;
a private modification area accessed by the groups of isolated operating system processes for supporting the view; and
a second set of names from the namespace used for direct access of files by the groups of isolated operating system processes in the private modification area,
wherein the private modification area is formed as a journaled transaction, and
wherein the private modification area stores an incremental difference between the files in the private modification area and the common template file system tree.
2. The system of claim 1 , further comprising multiple views inside the common template file system tree;
wherein the namespace includes a third set of names associated with the groups of isolated operating system processes for use in establishing the multiple views,
wherein read-only access to a file associated with the groups of isolated operating system processes using the third set of names utilizes the common template file system tree, and
wherein, if the same files are opened from different views of the common template file system tree using a name from the third set of names, these files are cached only once in an operating system file buffer.
3. The system of claim 2 , wherein a standard file system interface is utilized to enable file access that is transparent for the groups of isolated operating system processes.
4. The system of claim 2 , wherein if a file already exists in the private modification area, the read access is performed on that file.
5. The system of claim 2 , further comprising a transaction log, wherein contents of the transaction log corresponds to contents of the private modification area.
6. The system of claim 5 , wherein the contents of the transaction log is kept up to date and the transaction log is used for online replication.
7. The system of claim 2 , wherein a standard file system is utilized to enable file access that is transparent for any operating system process.
8. The system of claim 1 , further comprising multiple views inside a file system name space to the common template file system tree;
wherein a view inside a file system namespace to the common template file system tree includes a third set of names of the namespace associated with a virtual environment for use in establishing the multiple views;
wherein the common template file system tree is utilized by the groups of isolated operating system processes using the third set of names; and
wherein all files opened for write access are copied from the common template file system tree into the private modification area for performing the write access on the copy of the file.
9. The system of claim 8 , wherein a standard file system is utilized to enable file access that is transparent for the groups of isolated operating system processes.
10. The system of claim 8 , wherein:
if a file already exists in the private modification area, the write access is performed on that file, and
if the file does not exist in the private modification area, then the file is copied to the private modification area prior to the write access.
11. The system of claim 8 , wherein, if a file has been removed from the private modification area, a user is informed that the file does not exist.
12. A method of managing a computer file system comprising:
establishing a namespace;
launching a plurality of groups of isolated operating system processes;
establishing a first set of names from the namespace for direct access of the common template file system tree by the virtual environments;
establishing a common template file system tree for concurrent access by the groups of isolated operating system processes of the groups of isolated operating system processes using a view to the common template file system tree;
establishing a second set of names from the namespace for direct access of contents of files in a private modification area by the groups of isolated operating system processes;
accessing the private modification area by the isolated groups of operating system processes of the groups of isolated operating system processes; and
providing accessed files from the private modification area to users of the groups of isolated operating system processes,
wherein the private modification area is formed as a journaled transaction, and
wherein the private modification area stores an incremental difference between user files and the common template file system tree,
wherein, for shared files, only one copy of each shared file is cached by the operating system in memory, for all instances of processes using the shared files.
13. The method of claim 12 , further comprising establishing multiple views inside the common template file system tree,
wherein the namespace includes a third set of names associated with the groups of isolated operating system processes for use in establishing the multiple views,
wherein read-only access to a file associated with the groups of isolated operating system processes using the third set of names utilizes the common template file system tree, and
wherein, if the same files are opened from different views of the common template file system tree using a name from the third set of names, these files are cached only once in an operating system file buffer.
14. The method as defined in claim 13 , wherein if a file already exists in the private modification area, the read access is performed on that file.
15. The method of claim 12 , wherein a standard file system is utilized to enable file access that is transparent for the groups of isolated operating system processes.
16. The method of claim 12 , further comprising generating a transaction log, wherein contents of the transaction corresponds to contents of the private modification area.
17. The method as defined in claim 16 , further comprising keeping the transaction log up to date for online replication.
18. The method of claim 12 , wherein a standard file system is utilized to enable file access that is transparent for any operating system process.
19. The method of claim 12 , further comprising performing a write access on a file,
wherein the namespace includes a third set of names associated with the groups of isolated operating system processes, and
wherein all files opened for write access are copied from the common template file system tree into the private modification area prior to performing the write access.
20. The method of claim 19 , wherein:
if a file already exists in the private modification area, performing the write access on that file, and
if the file does not exist in the private modification area, copying the file to the private modification area prior to performing the write access on that file.
21. The method of claim 19 , wherein, if a file has been removed from the private modification area, informing a user that the file does not exist.
22. A computer useable medium having computer program logic stored thereon for executing on a processor and managing a computer file system, the computer program logic comprising:
computer program code means for establishing a namespace;
computer program code means for launching a plurality of groups of isolated operating system processes;
computer program code means for establishing a first set of names from the namespace for direct access of the common template file system tree by the virtual environments;
computer program code means for establishing a common template file system tree for concurrent access by the groups of isolated operating system processes using a view to the common template file system tree;
computer program code means for establishing a second set of names from the namespace for direct access of files in a private modification area by the groups of isolated operating system processes;
computer program code means for accessing the private modification area by the isolated operating system processes; and
computer program code means for providing accessed contents of files from the private modification area to users of the groups of isolated operating system processes,
wherein the private modification area is formed as a journaled transaction, and
wherein the private modification area stores an incremental difference between the files in the private modification area and the common template file system tree.
23. The computer useable medium of claim 22 , further comprising computer program code means for establishing multiple views inside the common template file system tree;
wherein the namespace includes a third set of names associated with the groups of isolated operating system processes for use in establishing the multiple views;
wherein read-only access to a file associated with the groups of isolated operating system processes using the third set of names utilizes the common template file system tree; and
wherein, if the same files are opened from different views of the common template file system tree using a name from the third set of names, these files are cached only once in an operating system file buffer.
24. The computer useable medium of claim 23 , wherein if a file already exists in the private modification area, the read access is performed on that file.
25. The computer useable medium of claim 22 , wherein a standard file system is utilized to enable file access that is transparent for the operating system.
26. The computer useable medium of claim 22 , further comprising computer program code means for generating a transaction log, wherein contents of the transaction corresponds to contents of the private modification area.
27. The computer useable medium of claim 26 , further comprising computer program code means for keeping the transaction log up to date for online replication.
28. The computer useable medium of claim 22 , wherein a standard file system is utilized to enable file access that is transparent for the operating system processes.
29. The computer useable medium of claim 22 , further comprising computer program code means for performing a write access on a file,
wherein the namespace includes a third set of names associated with the operating system processes,
wherein the common template file system tree is utilized by the groups of isolated operating system processes, and
wherein all files opened for write access are copied from the common template file system tree into the private modification area prior to performing the write access.
30. The computer useable medium of claim 29 , further comprising:
if a file already exists in the private modification area, computer program code means for performing the write access on that file, and
if the file does not exist in the private modification area, computer program code means for copying the file to the private modification area prior to performing the write access on that file.
31. The computer useable medium of claim 29 , wherein, if a file has been removed from the private modification area, informing a user that the file does not exist.
32. The computer useable medium of claim 22 , wherein a standard file system is utilized to enable file access that is transparent for any operating system process.
33. The system of claim 1 , wherein the system uses the first set of names from the namespace for direct access, using only a single redirect, of the common template file system tree by the groups of isolated operating system processes, and
wherein the system uses the second set of names from the namespace for direct access of files only in the private modification area and without a redirect to outside the private modification area.
34. A computer data storage file system comprising:
a storage device;
a namespace;
a plurality of groups of isolated operating system processes;
a common template file system tree stored on the storage device and concurrently accessed by the groups of isolated of operating system processes of the groups of isolated operating system processes using a view to the common template file system tree;
a first set of names from the namespace for direct access, using only a single redirect, of the common template file system tree by the groups of isolated operating system processes;
a private modification area accessed by the groups of isolated operating system processes for supporting the view; and
a second set of names from the namespace used for direct access of files only in the private modification area and without a redirect to outside the private modification area, by the groups of isolated operating system processes.
35. The system of claim 34 , wherein the private modification area is formed as a journaled transaction,
wherein the private modification area stores an incremental difference between user files and the common template file system tree,
wherein, for shared files, only one copy of each shared file is cached by the operating system in memory, for all instances of processes using the shared files.