Granted Patent B1

US 9,928,907 · App. 15/418,178 · Granted Mar 27, 2018

Abstract

Claims (49)

1. A storage device comprising:

a cross-point non-volatile memory (NVM) device that includes a first subset of cells, wherein cells of the first subset of cells share a bitline or a wordline, wherein for each adjacent pair of cells from the first subset of cells, there is at least one respective buffer cell that is between the respective cells of the adjacent pair of cells; and

a control module, the control module configured to:

receive a set of I/O operations; and

execute a first subset of the set of I/O operations in parallel across the first subset of cells, wherein I/O operations of the first subset of I/O operations are not executed at the respective buffer cells, wherein the at least one respective buffer cell is configured to not be programmed or erased when the first subset of the set of I/O operations are executed.

2. The storage device of claim 1 , wherein the control module is further configured to determine that a sufficient power budget exists to execute the first subset of the set of I/O operations in parallel based on voltage drop and current for performing the set of I/O operations on the first subset of cells.

3. The storage device of claim 1 , wherein each cell of the first subset of cells share both a respective wordline with at least a respective first cell of the first subset of cells and a respective bitline with at least a respective second cell of the first subset of cells.

4. The storage device of claim 1 , wherein the control module is further configured to:

determine a second subset of the set of I/O operations that can be executed in parallel with the first subset of I/O operations by analyzing chip power, voltage drop, and current requirements of the set of I/O operations; and

execute the second subset of I/O operations in parallel across a second subset of cells of the cross-point memory component as the first subset of I/O operations is executed in parallel across the first subset of cells, wherein:

cells of the second subset of cells share a wordline or bitline, and for each adjacent pair of cells from the second subset of cells, there is at least one respective buffer cell that is both not undergoing an I/O operation of the second subset of I/O operations and is between the respective cells of the adjacent pair of cells.

5. The storage device of claim 4 , wherein the first subset of cells and the second subset of cells includes the same number of cells.

6. The storage device of claim 1 , wherein the control module is further configured to determine a number of tiles of the cross-point NVM device by analyzing chip power of the cross-point NVM device, wherein the cross-point NVM device is configured to execute a subset of I/O operations at a subset of cells at each tile of the number of tiles in parallel, wherein the subset of cells includes the same number of cells as the first subset of cells.

7. The storage device of claim 1 , wherein the control module is further configured to:

determine that an amount of I/O operations within the set of I/O operations satisfies a threshold;

determine that the first subset of the set of I/O operations can be executed in parallel in response to the amount of I/O operations satisfying the threshold; and

wherein executing the first subset of I/O operations in parallel across the first subset of cells is in response to the amount of I/O operations satisfying the threshold.

8. The storage device of claim 1 , wherein the set of I/O operations comprises a first set of write operations, wherein the control module is further configured to:

receive a second set of I/O operations for the cross-point NVM device;

determine a threshold write current value based on voltage drop of the cross-point NVM device;

determine a number of cells in a second subset of cells based on the threshold write current; and

execute a subset of the second set of I/O operations across the second subset of cells, wherein cells of the second subset of cells share a bitline or a wordline, wherein for each adjacent pair of cells from the second subset of cells, there is at least one respective buffer cell that is both not undergoing an I/O operation of the second subset of I/O operations and is between the respective cells of the adjacent pair of cells.

9. The storage device of claim 1 , wherein the set of I/O operations relate to a set of cells that are physically clustered in a proximate location on the cross-point memory component, and the set of cells include the first subset of cells.

10. The storage device of claim 1 , wherein the NVM device includes the control module.

11. The storage device of claim 1 , wherein the storage device further comprises a controller, wherein the controller includes the control module.

12. A method of managing read or write operations to a cross-point non-volatile memory (NVM) device, the method comprising:

receiving a set of read or write operations for the cross-point NVM device, wherein the cross-point NVM device includes a plurality of cells grouped into a plurality of sets of cells, and wherein, for each set of cells, cells of the respective set of cells share a bitline or a wordline of the cross-point NVM device; and

executing a subset of the set of read or write operations in parallel across a set of cells of the plurality of sets of cells, wherein for adjacent pairs of cells from the set of cells there is at least one respective buffer cell that is both not undergoing a read or write operation of the subset of read or write operations and is between the respective cells of the adjacent pair of cells.

13. The method of claim 12 , the method further comprising determining that a sufficient power budget exists to execute a first subset of the set of read or write operations in parallel based on voltage drop and current for performing the set of I/O operations on a first subset of cells.

14. The method of claim 13 , the method further comprising:

determining a second subset of the set of read or write operations that can be executed in parallel with the first subset of read or write operations by analyzing chip power, voltage drop, and current requirements of the set of read or write operations; and

executing the second subset of read or write operations in parallel across a second subset of cells of the cross-point memory component as the first subset of read or write operations is executed in parallel across the first subset of cells, wherein:

cells of the second subset of cells share a wordline or bitline, and

for each adjacent pair of cells from the second subset of cells, there is at least one respective buffer cell that is both not undergoing a read or write operation of the second subset of read or write operations and is between the respective cells of the adjacent pair of cells.

15. The method of claim 14 , wherein the first subset of cells and the second subset of cells includes the same number of cells.

16. The method of claim 12 , wherein the set of read or write operations relate to cells that are physically clustered in a proximate location on the cross-point memory component, and the cells include the set of cells.

17. The method of claim 12 , the method further comprising:

determining that an amount of read or write operations within the set of read or write operations satisfies a threshold;

determining that the subset of the set of read or write operations can be executed in parallel in response to the amount of read or write operations satisfying the threshold; and

wherein executing the subset of read or write operations in parallel across the set of cells is in response to the amount of read or write operations satisfying the threshold.

18. The method of claim 12 , wherein the set of read or write operations comprise a first set of write operations, the method further comprising:

receiving a second set of read or write operations for the cross-point NVM device;

determining a threshold write current value based on voltage drop of the cross-point NVM device;

determining a number of cells in a second subset of cells based on the threshold write current; and

executing a subset of the second set of read or write operations across the second subset of cells, wherein cells of the second subset of cells share a bitline or a wordline, wherein for each adjacent pair of cells from the second subset of cells, there is at least one respective buffer cell that is both not undergoing a read or write operation of the second subset of read or write operations and is between the respective cells of the adjacent pair of cells.

19. A system comprising:

means for receiving a set of I/O operations for a cross-point non-volatile memory (NVM) device; and

means for executing a first subset of the set of I/O operations in parallel across a first subset of cells of the cross-point NVM device, wherein cells of the first subset of cells share a bitline or a wordline, wherein for each adjacent pair of cells from the first subset of cells, there is at least one respective buffer cell that is both not undergoing an I/O operation of the first subset of I/O operations and is between the respective cells of the adjacent pair of cells.

20. The system of claim 19 , further comprising means for determining that a sufficient power budget exists to execute the first subset of the set of I/O operations in parallel based on voltage drop and current for performing the set of I/O operations on the first subset of cells.

Assignments (10)