Systems and methods for file level prioritization during multi-object data restores
Methods and systems for file level prioritization during a data restore operation are disclosed. According to some embodiments, in response to a restore request to restore one or more files and for each file, the method includes retrieving file information and a critical value associated with the file, and calculating a Euclidean distance of the file to a consecutive file based on the critical value. The method further includes sorting the file(s) based on the calculated Euclidean distance of each file. The method further includes restoring the sorted file(s) at a target site.
1. A computer-implemented method for file level prioritization performed by a data deduplication system during a data recovery operation, the method comprising:
receiving a restore request to restore a plurality of files;
in response to the restore request, for each file among the plurality of files,
retrieving, from a thesis file, a filename of the file and a critical value associated with the filename, wherein the critical value indicates a file criticality of the file, and the thesis file comprises filenames of the plurality of files and corresponding critical values associated with filenames;
determining a priority weighted value of the file based on the critical value; and
determining a Euclidean distance of the file to a consecutive file based on the critical value associated with the filename and the priority weighted value;
sorting the plurality of files by selecting a nearest critical file based on the calculated Euclidean distance of each file; and
restoring the sorted files at a target device.
2. The method of claim 1 , further comprising prior to retrieving the filename of the file and the critical value associated with the filename, receiving the thesis file comprising the filenames of the plurality of files and the corresponding critical values associated with the filenames.
3. The method of claim 1 , wherein determining the Euclidean distance of the file to the consecutive file comprises taking a dot product of the critical value and a sequence of the file in a queue.
4. The method of claim 1 , wherein restoring the sorted files at the target device comprises restoring a nearest and most critical file in accordance with the calculated Euclidean distance of each file.
5. The method of claim 1 , further comprising allocating data streams according to nearest and critical files from the plurality of files.
6. The method of claim 2 , wherein the thesis file is received on an on-demand basis.
7. The method of claim 1 , further comprising routing the sorted files to a target buffer within available data streams for read operations.
8. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations, the operations comprising:
receiving a restore request to restore a plurality of files;
in response to the restore request, for each file among the plurality of files,
retrieving, from a thesis file, a filename of the file and a critical value associated with the filename, wherein the critical value indicates a file criticality of the file, and the thesis file comprises filenames of the plurality of files and corresponding critical values associated with filenames;
determining a priority weighted value of the file based on the critical value; and
determining a Euclidean distance of the file to a consecutive file based on the critical value associated with the filename and the priority weighted value;
sorting the plurality of files by selecting a nearest critical file based on the calculated Euclidean distance of each file; and
restoring the sorted files at a target device.
9. The non-transitory machine-readable medium of claim 8 , wherein the operations further comprise prior to retrieving the filename of the file and the critical value associated with the filename, receiving the thesis file comprising the filenames of the plurality of files and the corresponding critical values associated with the filenames.
10. The non-transitory machine-readable medium of claim 8 , wherein determining the Euclidean distance of the file to the consecutive file comprises taking a dot product of the critical value and a sequence of the file in a queue.
11. The non-transitory machine-readable medium of claim 8 , wherein restoring the sorted files at the target device comprises restoring a nearest and most critical file in accordance with the calculated Euclidean distance of each file.
12. The non-transitory machine-readable medium of claim 8 , wherein the operations further comprise allocating data streams according to nearest and critical files from the plurality of files.
13. The non-transitory machine-readable medium of claim 9 , wherein the thesis file is received on an on-demand basis.
14. The non-transitory machine-readable medium of claim 8 , wherein the operations further comprise routing the sorted files to a target buffer within available data streams for read operations.
15. A data processing system, comprising:
a processor; and
a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations, the operations including:
receiving a restore request to restore a plurality of files;
in response to the restore request, for each file among the plurality of files,
retrieving, from a thesis file, a filename of the file and a critical value associated with the filename, wherein the critical value indicates a file criticality of the file, and the thesis file comprises filenames of the plurality of files and corresponding critical values associated with filenames;
determining a priority weighted value of the file based on the critical value; and
determining a Euclidean distance of the file to a consecutive file based on the critical value associated with the filename and the priority weighted value;
sorting the plurality of files by selecting a nearest critical file based on the calculated Euclidean distance of each file; and
restoring the sorted files at a target device.
16. The data processing system of claim 15 , wherein the operations further include prior to retrieving the filename of the file and the critical value associated with the filename, receiving the thesis file comprising the filenames of the plurality of files and the corresponding critical values associated with the filenames.
17. The data processing system of claim 15 , wherein determining the Euclidean distance of the file to the consecutive file comprises taking a dot product of the critical value and a sequence of the file in a queue.
18. The data processing system of claim 15 , wherein restoring the sorted files at the target device comprises restoring a nearest and most critical file in accordance with the calculated Euclidean distance of each file.
19. The data processing system of claim 15 , wherein the operations further include allocating data streams according to nearest and critical files from the plurality of files.