System and method to improve input output command latency by dynamic size logical to physical caching
A method and apparatus are provided to divide a logical to physical table into multiple parts, one part in a first fast memory and a second part in a second non-volatile memory, wherein an algorithm may be used in the division.
1. A method for operating an apparatus, comprising:
obtaining a logical to physical table for the apparatus, the logical to physical mapping logical entities to physical entities;
analyzing data in the logical to physical table;
choosing a parameter to determine a first portion and a second portion of a dynamic logical to physical table;
sorting the logical to physical table according to the chosen parameter; and
apportioning a size of the first portion and the second portion of the dynamic logical to physical table based on the parameter.
2. The method according to claim 1 , wherein the first portion of the logical to physical table is placed in a fast memory.
3. The method according to claim 2 , wherein the fast memory is a volatile memory.
4. The method according to claim 1 , wherein the second portion of the logical to physical table is placed in a slow memory.
5. The method according to claim 4 , wherein the slow memory is a DDR memory.
6. The method according to claim 4 , wherein the slow memory is a NAND flash memory.
7. The method according to claim 1 , wherein the parameter is a namespace.
8. The method according to claim 1 , further comprising:
saving the first portion to a fast memory and saving the second portion to a slow memory.
9. An arrangement for augmenting a logical to physical table, comprising:
means for analyzing data in the logical to physical table;
means for choosing a parameter to determine a first portion and a second portion of a dynamic logical to physical table;
means for sorting the logical to physical table according to the chosen parameter; and
means for apportioning a size of the first portion and the second portion of the dynamic logical to physical table based on the parameter.
10. The arrangement according to claim 9 , further comprising:
means for saving the first portion to a fast memory and saving the second portion to a slow memory.
11. A method for operating an apparatus, comprising:
obtaining a logical to physical table;
choosing a parameter to determine a first portion and a second portion of the logical to physical table;
apportioning the first portion and the second portion of the logical to physical table based upon the chosen parameter;
performing operations based upon the logical to physical table;
determining if the performed operations with the logical to physical table include the second portion of the logical to physical table; and
augmenting the logical to physical table, wherein data in the second portion of the logical to physical table used in operations are placed in the first portion of the logical to physical table.
12. The method according to claim 11 , further comprising:
saving the logical to physical table after the apportioning the first portion and the second portion of the logical to physical table based upon the chosen parameter.
13. The method according to claim 11 , further comprising:
saving the logical to physical table after the augmenting the logical to physical table, wherein data in the second portion of the logical to physical table used in operations are placed in the first portion of the logical to physical table.
14. The method according to claim 11 , wherein the first portion of the logical to physical table is placed in a fast memory.
15. The method according to claim 14 , wherein the fast memory is a volatile memory.
16. The method according to claim 11 , wherein the second portion of the logical to physical table is placed in a slow memory.
17. The method according to claim 16 , wherein the slow memory is a DDR memory.
18. The method according to claim 16 , wherein the slow memory is a NAND flash memory.
19. The method according to claim 11 , wherein the parameter is a namespace.
20. The method according to claim 11 , wherein the parameter is a type and amount of input output workload for a namespace.
21. The method according to claim 20 , wherein an algorithm is used to determine an amount of first portion capacity and a second portion capacity.
22. An arrangement for augmenting a logical to physical table, comprising:
means for choosing a parameter to determine a first portion and a second portion of the logical to physical table;
means for apportioning the first portion and the second portion of the logical to physical table based upon the chosen parameter;
means for performing operations based upon the logical to physical table;
means for determining if the performed operations with the logical to physical table include the second portion of the logical to physical table; and
means for augmenting the logical to physical table, wherein data in the second portion of the logical to physical table used in operations are placed in the first portion of the logical to physical table.
23. The arrangement according to claim 22 , further comprising:
means for saving the logical to physical table after the augmenting the logical to physical table, wherein data in the second portion of the logical to physical table used in operations are placed in the first portion of the logical to physical table.
24. A method for operating a solid state drive, comprising:
obtaining a logical to physical table, the logical to physical table tabulating locations of data stored on the solid state drive;
choosing a parameter to divide the logical to physical table into at least a first portion and a second portion;
dividing the logical to physical table into the first portion and the second portion based upon the chosen parameter; and
storing the first portion of the logical to physical table in a first memory and the second portion of the logical to physical table in a second memory.
25. The method according to claim 24 , wherein the first memory is a volatile memory and the second memory is a non-volatile memory.
26. The method according to claim 24 , wherein the parameter is used by an algorithm for the dividing the logical to physical table.