COMPUTING SYSTEM USING NONVOLATILE MEMORY AS MAIN MEMORY AND METHOD FOR MANAGING THE SAME

• US 20140173189A1
• Filed: 12/10/2013
• Published: 06/19/2014
• Est. Priority Date: 12/18/2012
• Status: Abandoned Application

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Hwajin Jung, Young-Jin Park
• Inventors: PARK, YOUNG-JIN, JUNG, HWAJIN
• Application Number:
  US14/101,379
• Publication Number:
  US 20140173189A1
• Time in Patent Office: Days
• US Class Current: 711/103
• CPC Class Codes: G06F 12/0238 Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
• CPC Class Codes: G06F 2212/202 Non-volatile memory
• CPC Class Codes: G06F 9/461 Saving or restoring of program or task context
• CPC Class Codes: G06F 9/4806 Task transfer initiation or dispatching

Abstract

A method of managing data of a computing system is provided, where the computing system uses a nonvolatile memory as a main memory. The method includes loading a process into the nonvolatile memory in response to a first run request, freezing the process loaded into the nonvolatile memory in response to an exit request of the process, and activating the process frozen in the nonvolatile memory in response to a second run request of the process. Freezing the process releases control of the process without deleting the process loaded into the nonvolatile memory.

Claims

1. A method of managing data of a computing system, which uses a nonvolatile memory as a main memory, the method comprising:

loading a process into the nonvolatile memory in response to a first run request;

freezing the process loaded into the nonvolatile memory in response to an exit request of the process; and

activating the process frozen in the nonvolatile memory in response to a second run request of the process,

wherein freezing the process releases control of the process without deleting the process loaded into the nonvolatile memory.

2. The method of claim 1, wherein freezing the process retains a memory area of the nonvolatile memory corresponding to the process without releasing.

3. The method of claim 1, further comprising:

adding an address allocated to the process to a run process table in response to the first run request.

4. The method of claim 3, further comprising:

moving the address of the process registered in the run process table to an exit process table in response to the exit request.

5. The method of claim 4, wherein moving the address comprises:

selecting a priority of the process referring to a priority table; and

moving the address of the process to the exit process table according to the selected priority.

6. The method of claim 5, wherein the priority is selected according to information on a time when the process is most recently used or a run frequency of the process.

7. The method of claim 6, wherein the priority of the process increases in proportion to an increase in a run frequency of the process.

8. The method of claim 6, wherein the priority of the process increases in proportion to an increase in an exit time of the particular process.

9. The method of claim 6, wherein each of the run process table and the exit process table is configured to store a run frequency of each process.

10. The method of claim 5, wherein moving the address of the process to the exit process table according to the selected priority comprises:

comparing priorities of processes previously registered in the exit process table with the selected priority.

11. The method of claim 4, wherein when a release request is generated, memory areas of processes registered in the exit process table are released in an order from a process having a lower priority to a process having a higher priority.

12. The method of claim 11, wherein an address of a process corresponding to a released memory area is removed from the exit process table.

13. The method of claim 1, further comprising:

releasing a memory area corresponding to a process frozen in the nonvolatile memory when free space of the nonvolatile memory is not sufficient for loading the process.

14. The method of claim 1, wherein memory areas corresponding to processes loaded into the nonvolatile memory or frozen processes are not released even at power-on or power-off.

15. A computing system, comprising:

a nonvolatile storage;

a nonvolatile main memory; and

a processor configured to load a process into the nonvolatile main memory from the nonvolatile storage in response to a first run request, to freeze the process loaded into the nonvolatile main memory in response to an exit request of the process, and to activate the process frozen in the nonvolatile main memory in response to a second run request of the process,

wherein the freezing releases control of the process without deleting of the process loaded into the nonvolatile main memory; and

wherein the processor is further configured to skip an operation of loading the process into the nonvolatile main memory from the nonvolatile storage in response to the second run request.

16. The computing system of claim 15, wherein the processor comprises a process manager having a run process table for storing information on the process in response to the first run request and an exit process table for storing information on the process transferred from the run process table in response to the exit request.

17. The computing system of claim 16, wherein the run process table further stores information on the process transferred from the exit process table in response to the second run request.

18. A processor in a computing system including a main memory, the processor comprising:

a core configured to control processes of the computing system;

a main memory unit configured to manage the main memory under control of the core; and

a process manager configured to manage information on processes run and exited by the core, the process manager comprising a run process table for storing information on processes in response to run requests and an exit process table for storing information on processes transferred from the run process table in response to exit requests, the process manager determining whether a process of a run request, received by the core, is cached using the exit process table,

wherein, when the process manager determines that the process is cached, one of the core and the main memory unit activates a memory area in which the process is stored, and moves information regarding the process from the exit process table to the run process table of the process manager.

19. The process of claim 18, wherein, when the process manager determines that the process is not cached, one of the core and the main memory unit allocates a memory area of the main memory for the process, and adds information regarding the process to the run process table of the process manager.

20. The process of claim 19, wherein the information regarding the process added to the run process table comprises process name and process address, and the information regarding the process included in the exit process table comprises process name, process address and priority in relation to at least one other process.