Granted Patent B2

US 8,209,303 · App. 12/511,624 · Granted Jun 26, 2012

Abstract

Claims (49)

1. A system for implementing a reliable persistent randomly accessible compressed data stream, the system comprising:

a computer-implemented journaled file system comprising:

a first file for storing a series of independently compressed blocks of a data stream;

a second file for storing a series of indexes corresponding to the compressed blocks, wherein each one of the indexes comprises a byte offset into the first file of the corresponding compressed block; and

a third file for storing a chunk of data from the data stream before it is compressed and written to the first file;

a writer module for writing a first chunk of data from the data stream before it is compressed to the third file and writing a first index to the second file, and for writing a second chunk of data from the data stream before it is compressed to the third file and writing a second index to the second file; and

a compressor module for compressing the first chunk of data from the third file and writing it to the first file, and for compressing the second chunk of data from the third file and writing it to the first file,

wherein each of the indexes comprises an offset into the first file of the corresponding block.

2. The system of claim 1 wherein each of the indexes is an unsigned integer.

3. The system of claim 1 further comprising:

a reader module for reading a block of data from a location in the first file; and

a decompressor module for decompressing the read block of data for presentation to a user.

4. The system of claim 3 wherein the reader module determines a block index and offset for the location using the second file.

5. The system of claim 1 wherein the third file has associated therewith a zero padded hexadecimal number corresponding to a block number associated with the data chunk being staged therein.

6. The system of claim 1 wherein the journaled file system supports ordered journaling and ordered write guarantees.

7. A method for implementing a reliable persistent randomly accessible compressed data stream, the method comprising:

creating an initially empty STAGE file, BLOCKS file, and INDEX file;

writing a first chunk of data from a data stream to the STAGE file before the first chunk of data is compressed and written to the BLOCKS file;

independently compressing the first chunk of data written to the STAGE file and writing it as a compressed first block to the BLOCKS file;

writing a first index corresponding to the compressed first block to the INDEX file;

writing a second chunk of data from the data stream to the STAGE file before the second chunk of data is compressed and written to the BLOCKS file;

independently compressing the second chunk of data written to the STAGE file and writing it as a compressed second block to the BLOCKS file; and

writing a second index corresponding to the compressed second block to the INDEX file,

wherein each of the indexes comprises a byte offset into the BLOCKS file of the corresponding compressed block.

8. The method of claim 7 wherein each of the indexes is an unsigned integer.

9. The method of claim 7 further comprising:

reading a block of data from a location in the BLOCKS file; and

decompressing the read block of data for presentation to a user.

10. The method of claim 9 wherein reading includes determining a block index and offset for the location using the INDEX file.

11. The method of claim 7 wherein the BLOCKS file contains a number of blocks of data,

wherein the INDEX file contains a number of indexes,

wherein the number of blocks of data and the number of indexes are the same.

12. The method of claim 7 wherein the STAGE file has associated therewith a zero padded hexadecimal number corresponding to a block number associated with the data chunk being staged therein.

13. A method for implementing a reliable persistent randomly accessible compressed data stream, the method comprising:

accumulating first data from a data stream in a STAGE file;

responsive to accumulation of a first predetermined amount of first data in the STAGE file, compressing the accumulated first data into a first block of data;

writing the compressed first block of data to a BLOCKS file at a first offset;

writing a first index indicating the first offset of the compressed first block to an INDEX file;

after writing the compressed first block of data and the first index, accumulating second data from the data stream in the STAGE file;

responsive to accumulation of a second predetermined amount of second data in the STAGE file, compressing the accumulated second data into a second block of data;

writing the compressed second block of data to the BLOCKS file at a second offset;

writing a second index indicating the second offset of the compressed second block to the INDEX file.

14. The method of claim 13 wherein the second offset is a next available location in the BLOCKS file after the compressed first block of data.

15. The method of claim 13 further comprising, responsive to a request to read a block of data from a location in the BLOCKS file, determining a block index and offset for the location using the INDEX file.

16. The method of claim 13 wherein each of the indexes is an unsigned integer.

17. The method of claim 13 wherein the BLOCKS file contains blocks of data and wherein the INDEX file contains indexes corresponding in order to the blocks, such that each index comprises an offset into the BLOCKS file of the corresponding block.

18. The method of claim 13 wherein the STAGE file has associated therewith a zero padded hexadecimal number corresponding to a block number associated with the data accumulating therein.

19. The method of claim 13 wherein the STAGE file is recycled.

20. The method of claim 13 wherein the first predetermined amount and the second predetermined amount are the same.

Assignments (16)