Overwriting compressed data extents
View Patent ↗A technique for overwriting compressed data tests whether new data compressed with a first compression procedure fits within spaces provided for previous data. If the compressed new data does not fit, the technique compresses the new data using a second compression procedure. Assuming the second compression procedure reduces the compressed size of the new data to fit the available space, the technique stores the new data in the same location as the previous data. In this manner, overwrites can be accommodated in place without the need to create new mapping metadata.
1. A method of overwriting compressed data in a data storage system, the method comprising:
receiving an IO (Input/Output) request specifying a write of new data to a specified logical address of a file within a file system, the file already storing previous data at the specified logical address such that the write of the new data is an overwrite;
compressing the new data using a first compression procedure;
performing a testing operation configured to generate a first result when the compressed new data fits within a space currently occupied by the previous data and to produce a second result otherwise; and
in response to the testing operation producing the second result, (i) compressing the new data using a second compression procedure and (ii) storing the new data as compressed using the second compression procedure within the space currently occupied by the previous data.
2. The method of claim 1 , wherein, prior to storing the new data, the previous data was stored along with other compressed data in a segment of blocks laid out contiguously in a physical address space of the file system, and wherein the method further comprises, when storing the new data:
identifying, within the segment of blocks, a compression header of the previous data; and
updating the compression header to store an identifier of the second compression procedure.
3. The method of claim 2 , wherein the compression header of the previous data also includes a field for storing a compressed length of the previous data, and wherein the method further comprises:
updating the field for storing the compressed length to a length of the new data as compressed using the second compression procedure.
4. The method of claim 3 , wherein the file system includes a set of metadata structures that map the logical address of the previous data to the segment of blocks, and wherein storing the new data is performed without changing any of the set of metadata structures.
5. The method of claim 3 ,
wherein the file system includes a set of metadata structures that map the logical address of the previous data to the segment of blocks,
wherein the set of metadata structures includes a virtual block map (VBM) that stores an extent list and a pointer to the segment of blocks,
wherein the extent list includes (i) the logical address in the file of the previous data and (ii) a version of the compressed length of the previous data, and
wherein storing the new data is performed without changing the version of the length of the previous data in the extent list.
6. The method of claim 5 , further comprising:
monitoring CPU (Central Processing Unit) activity of a CPU in the data storage system; and
performing a CPU activity test configured to produce a GO result in response to CPU activity falling below a predetermined threshold and to produce a NO-GO result otherwise,
wherein compressing the new data using the second compression procedure is performed in response to the CPU activity test producing the GO result.
7. The method of claim 2 , further comprising:
receiving a second IO request specifying a write of second new data to a second logical address of the file within the file system, the file already storing previous second data at the second logical address;
compressing the second new data using the first compression procedure;
performing the testing operation on the compressed second new data; and
in response to the testing operation on the compressed second new data producing the second result, storing the compressed second new data at another location in the file system.
8. A data storage system, comprising control circuitry that includes a set of processing units coupled to memory, the control circuitry constructed and arranged to:
receive an IO (Input/Output) request specifying a write of new data to a specified logical address of a file within a file system, the file already storing previous data at the specified logical address such that the write of the new data is an overwrite;
compress the new data using a first compression procedure;
perform a testing operation configured to generate a first result when the compressed new data fits within a space currently occupied by the previous data and to produce a second result otherwise; and
in response to the testing operation producing the second result, (i) compress the new data using a second compression procedure and (ii) store the new data as compressed using the second compression procedure within the space currently occupied by the previous data.
9. The data storage system of claim 8 , wherein, prior to storing the new data, the previous data was stored along with other compressed data in a segment of blocks laid out contiguously in a physical address space of the file system, and wherein the control circuitry, when constructed and arranged to store the new data, is further constructed and arranged to:
identify, within the segment of blocks, a compression header of the previous data; and
update the compression header to store an identifier of the second compression procedure.
10. The data storage system of claim 9 , wherein the compression header of the previous data also includes a field for storing a compressed length of the previous data, and wherein the control circuitry is further constructed and arranged to:
update the field for storing the compressed length to a length of the new data as compressed using the second compression procedure.
11. The data storage system of claim 10 ,
wherein the file system includes a set of metadata structures that map the logical address of the previous data to the segment of blocks,
wherein the set of metadata structures includes a virtual block map (VBM) that stores an extent list and a pointer to the segment of blocks,
wherein the extent list includes (i) the logical address in the file of the previous data and (ii) a version of the compressed length of the previous data, and
wherein the control circuitry is constructed and arranged to store the new data without changing the version of the length of the previous data in the extent list.
12. The data storage system of claim 11 , wherein the control circuitry is further constructed and arranged to:
monitor CPU (Central Processing Unit) activity of a CPU in the data storage system; and
perform a CPU activity test configured to produce a GO result in response to CPU activity falling below a predetermined threshold and to produce a NO-GO result otherwise,
compress the new data using the second compression procedure in response to the CPU activity test producing the GO result.
13. The data storage system of claim 9 , wherein the control circuitry is further constructed and arranged to:
receive a second IO request specifying a write of second new data to a second logical address of the file within the file system, the file already storing previous second data at the second logical address;
compress the second new data using the first compression procedure;
perform the testing operation on the compressed second new data; and
in response to the testing operation on the compressed second new data producing the second result, store the compressed second new data at another location in the file system.
14. A computer program product including a set of non-transitory, computer-readable media having instructions which, when executed by control circuitry of a data storage system, cause the control circuitry to perform a method for overwriting compressed data, the method comprising:
receiving an IO (Input/Output) request specifying a write of new data to a specified logical address of a file within a file system, the file already storing previous data at the specified logical address such that the write of the new data is an overwrite;
compressing the new data using a first compression procedure;
performing a testing operation configured to generate a first result when the compressed new data fits within a space currently occupied by the previous data and to produce a second result otherwise; and
in response to the testing operation producing the second result, (i) compressing the new data using a second compression procedure and (ii) storing the new data as compressed using the second compression procedure within the space currently occupied by the previous data.
15. The computer program product of claim 14 , wherein, prior to storing the new data, the previous data was stored along with other compressed data in a segment of blocks laid out contiguously in a physical address space of the file system, and wherein the method further comprises, when storing the new data:
identifying, within the segment of blocks, a compression header of the previous data; and
updating the compression header to store an identifier of the second compression procedure.
16. The computer program product of claim 15 , wherein the compression header of the previous data also includes a field for storing a compressed length of the previous data, and wherein the method further comprises:
updating the field for storing the compressed length to a length of the new data as compressed using the second compression procedure.
17. The computer program product of claim 16 , wherein the file system includes a set of metadata structures that map the logical address of the previous data to the segment of blocks, and wherein storing the new data is performed without changing any of the set of metadata structures.
18. The computer program product of claim 16 ,
wherein the file system includes a set of metadata structures that map the logical address of the previous data to the segment of blocks,
wherein the set of metadata structures includes a virtual block map (VBM) that stores an extent list and a pointer to the segment of blocks,
wherein the extent list includes (i) the logical address in the file of the previous data and (ii) a version of the compressed length of the previous data, and
wherein storing the new data is performed without changing the version of the length of the previous data in the extent list.
19. The computer program product of claim 18 , wherein the method further comprises:
monitoring CPU (Central Processing Unit) activity of a CPU in the data storage system; and
performing a CPU activity test configured to produce a GO result in response to CPU activity falling below a predetermined threshold and to produce a NO-GO result otherwise,
wherein compressing the new data using the second compression procedure is performed in response to the CPU activity test producing the GO result.
20. The computer program product of claim 15 , wherein the method further comprises:
receiving a second IO request specifying a write of second new data to a second logical address of the file within the file system, the file already storing previous second data at the second logical address;
compressing the second new data using the first compression procedure;
performing the testing operation on the compressed second new data; and
in response to the testing operation on the compressed second new data producing the second result, storing the compressed second new data at another location in the file system.