Method of decreasing write amplification of NAND flash using a journal approach
A journaling approach is used to distribute data of different sizes between areas of a segment's log on a physical NAND flash erase block. The Main area contains large, contiguous extents of data, and the Journal area contains logical blocks of small data. An Updates area also contains updates that are pending. One disclosed embodiment includes storing a first file fragment associated with a file in a journal area of a log, where a size of the file fragment is less than a physical NAND flash page size limit, receiving a second file fragment associated with the file, combining the first file fragment and the second file fragment when a combined size of the fragments is equal to the physical NAND flash page size limit, storing the combined fragments in a main area of a second log, receiving an update associated with the combined fragments, and storing the update in an updates area of a third log.
1. A method of storing file fragments of growing files in a flash-based storage system, comprising:
storing a first file fragment associated with a file in a journal area of a first log in flash memory, wherein a size of the file fragment is less than a physical NAND flash page size limit;
receiving a second file fragment associated with the file;
combining the first file fragment and the second file fragment when a combined sized of the fragments is equal to the physical NAND flash page size limit;
storing the combined fragments in a main area of a second log in flash memory, wherein the second log is different from the first log;
receiving an update associated with the combined fragments; and
storing the update in an updates area of a third log in flash memory, wherein the third log is different from the first and second logs,
wherein each of the first, second and third logs comprises a respective journal area, a respective main area and a respective updates area.
2. The method of claim 1 , wherein a size of the update is less than the physical NAND flash page size limit.
3. The method of claim 1 , wherein the main area comprises content that is rarely updated, and the updates area and the journal area comprise content that is frequently updated.
4. The method of claim 1 , wherein the flash-based storage system comprises NAND flash.
5. The method of claim 1 , wherein the combined fragments are stored in a pre-allocated NAND flash page of the main area.
6. The method of claim 1 , further comprising updating metadata information associated with the first, second, and third logs.
7. An apparatus for storing file fragments of growing files in a flash-based storage system, comprising:
a flash memory device;
a main memory; and
a processor communicatively coupled to the flash memory device and the main memory, the processor configured to:
store a first file fragment associated with a file in a journal area of a first log in the flash memory device, wherein a size of the file fragment is less than a physical NAND flash page size limit;
receive a second file fragment associated with the file;
combine the first file fragment and the second file fragment when a combined size of the fragments is equal to the physical NAND flash page size limit;
store the combined fragments in a main area of a second log in the flash memory device, wherein the second log is different from the first log;
receive an update associated with the combined fragments; and
store the update in an updates area of a third log in the flash memory device, wherein the third log is different from the first and second logs,
wherein each of the first, second and third logs comprises a respective journal area, a respective main area and a respective updates area.
8. The apparatus of claim 7 , wherein the first, second, and third logs are constructed in the main memory and stored on the flash memory device.
9. The apparatus of claim 7 , wherein a size of the update is less than the physical NAND flash page size limit.
10. The apparatus of claim 7 , wherein the main area comprises content that is rarely updated, and the updates area and the journal area comprise content that is frequently updated.
11. The apparatus of claim 7 , wherein the flash-based storage system comprises NAND flash.
12. The apparatus of claim 7 , wherein the combined fragments are stored in a pre-allocated NAND flash page of the main area.
13. A computer program product tangibly embodied in a computer-readable storage device and comprising instructions that when executed by a processor perform a method for storing file fragments of growing files in a flash-based storage system, the method comprising:
storing a first file fragment associated with a file in a journal area of a first log in flash memory, wherein a size of the file fragment is less than a physical NAND flash page size limit;
receiving a second file fragment associated with the file;
combining the first file fragment and the second file fragment when a combined size of the fragments is equal to the physical NAND flash page size limit;
storing the combined fragments in a main area of a second log in flash memory, wherein the second log is different from the first log;
receiving an update associated with the combined fragments; and
storing the update in an updates area of a third log in flash memory, wherein the third log is different from the first and second logs,
wherein each of the first, second and third logs comprises a respective journal area, a respective main area and a respective updates area.
14. The computer program product of claim 13 , wherein a size of the update is less than the physical NAND flash page size limit.
15. The computer program product of claim 13 , wherein the main area comprises content that is rarely updated, and the updates area and the journal area comprise content that is frequently updated.
16. The computer program product of claim 13 , wherein the flash-based storage system comprises NAND flash.
17. The computer program product of claim 13 , wherein the combined fragments are stored in pre-allocated NAND flash page of the main area.
18. The computer program product of claim 13 , further comprising updating metadata information associated with the first, second, and third logs.