Performance of file system operations

1. A method of improving performance of file system operations, comprising:

copying dirty memory pages from a file system into a memory associated with a first hardware accelerator associated with a host;

compressing content of the memory pages by one or more cores of the first hardware accelerator;

compacting compressed data into a payload area of at least one log;

preparing metadata of the at least one log by the one or more cores of the first hardware accelerator;

flushing the at least one log into a storage device;

wherein the method further comprises:

identifying a location of the at least one log using a mapping table by the first hardware accelerator in response to a request of reading data in the at least one log from at least one physical sector;

accessing the at least one log by the first hardware accelerator based on the identified location;

parsing the at least one log and executing decompression operations by the first hardware accelerator to retrieve the requested data; and

delivering the requested data to a page cache of the file system.

2. The method of claim 1 , further comprising:

creating a mapping table by the first hardware accelerator, the mapping table configured to identify locations of logs in the storage device.

3. The method of claim 1 , further comprising:

identifying a location of the at least one log using a mapping table in response to a request of reading data in the at least one log from a plurality of physical sectors;

parsing the at least one log and executing decompression operations by a second hardware accelerator associated with the storage device;

in response to determining that a size of data extracted from the plurality of physical sectors is greater than a predetermined threshold, creating a new log by combining the data from the plurality of physical sectors using compression by the second hardware accelerator associated with the storage device; and

sending the new log to the first hardware accelerator.

4. The method of claim 3 , further comprising:

parsing the new log and executing decompression operations by the first hardware accelerator associated with the host; and

delivering the requested data to a page cache of the file system.

5. The method of claim 3 , wherein the creating a new log further comprises:

executing delta-decoding operations by the second hardware accelerator; and

creating the new log for delta-decoded data.

6. The method of claim 1 , wherein the at least one log further comprises a header depicting a structure of the payload area.

7. The method of claim 1 , further comprising:

processing a plurality of requests for log creation in parallel by the first hardware accelerator.

8. The method of claim 1 , further comprising:

executing data operations associated with different logs in a massively parallel manner.

9. A system, comprising:

at least one processor; and

at least one memory comprising computer-readable instructions that upon execution by the at least one processor cause the computing device to perform operations comprising:

copying dirty memory pages from a file system into a memory associated with a first hardware accelerator associated with a host;

compressing content of the memory pages by one or more cores of the first hardware accelerator;

compacting compressed data into a payload area of at least one log;

preparing metadata of the at least one log by the one or more cores of the first hardware accelerator;

flushing the at least one log into a storage device;

wherein the operations further comprise:

identifying a location of the at least one log using a mapping table by the first hardware accelerator in response to a request of reading data in the at least one log from at least one physical sector;

accessing the at least one log by the first hardware accelerator based on the identified location;

parsing the at least one log and executing decompression operations by the first hardware accelerator to retrieve the requested data; and

delivering the requested data to a page cache of the file system.

10. The system of claim 9 , the operations further comprising:

creating a mapping table by the first hardware accelerator, the mapping table configured to identify locations of logs in the storage device.

11. The system of claim 9 , the operations further comprising:

identifying a location of the at least one log using a mapping table in response to a request of reading data in the at least one log from a plurality of physical sectors;

parsing the at least one log and executing decompression operations by a second hardware accelerator associated with the storage device;

in response to determining that a size of data extracted from the plurality of physical sectors is greater than a predetermined threshold, creating a new log by combining the data from the plurality of physical sectors using compression by the second hardware accelerator associated with the storage device; and

sending the new log to the first hardware accelerator.

12. The system of claim 11 , the operations further comprising:

parsing the new log and executing decompression operations by the first hardware accelerator associated with the host; and

delivering the requested data to a page cache of the file system.

13. The system of claim 11 , wherein the creating a new log further comprises:

executing delta-decoding operations by the second hardware accelerator; and

creating the new log for delta-decoded data.

14. A non-transitory computer-readable storage medium, storing computer-readable instructions that upon execution by a processor cause the processor to implement operations, the operations comprising:

copying dirty memory pages from a file system into a memory associated with a first hardware accelerator associated with a host;

compressing content of the memory pages by one or more cores of the first hardware accelerator;

compacting compressed data into a payload area of at least one log;

preparing metadata of the at least one log by the one or more cores of the first hardware accelerator;

flushing the at least one log into a storage device;

wherein the operations further comprise:

identifying a location of the at least one log using a mapping table by the first hardware accelerator in response to a request of reading data in the at least one log from at least one physical sector;

accessing the at least one log by the first hardware accelerator based on the identified location;

parsing the at least one log and executing decompression operations by the first hardware accelerator to retrieve the requested data; and

delivering the requested data to a page cache of the file system.

15. The non-transitory computer-readable storage medium of claim 14 , the operations further comprising:

creating a mapping table by the first hardware accelerator, the mapping table configured to identify locations of logs in the storage device.

16. The non-transitory computer-readable storage medium of claim 14 , the operations further comprising:

identifying a location of the at least one log using a mapping table in response to a request of reading data in the at least one log from a plurality of physical sectors;

parsing the at least one log and executing decompression operations by a second hardware accelerator associated with the storage device;

in response to determining that a size of data extracted from the plurality of physical sectors is greater than a predetermined threshold, creating a new log by combining the data from the plurality of physical sectors using compression by the second hardware accelerator associated with the storage device; and

sending the new log to the first hardware accelerator.

17. The non-transitory computer-readable storage medium of claim 16 , the operations further comprising:

parsing the new log and executing decompression operations by the first hardware accelerator associated with the host; and

delivering the requested data to a page cache of the file system.