Catastrophic event memory backup system
A persistent memory unit for a computer system where the memory unit can detect a catastrophic event and automatically backup volatile memory into non-volatile memory. The memory unit can operate with a limited number of power inputs and detect the loss of power and then initiate a backup after the volatile memory of the memory unit has entered a stable self-refresh mode. The memory unit uses an on-board power management interface controller capable of redistributing power from an input power line and generating different power levels for different components on the memory unit.
1. A method of operation of a hybrid memory unit comprising a non-volatile memory and a volatile memory, the method comprising:
detecting a catastrophic event on an external host computing system that is operably connected to the hybrid memory unit;
upon detecting the catastrophic event, receiving power from an alternate power source;
executing a safety condition action for the volatile memory of the hybrid memory unit, wherein executing the safety condition action comprises initiating the safety condition action based on receiving a failure status at a registering clock driver of the hybrid memory unit;
generating a safety status based on completion of the safety condition action;
generating a trigger status based on the safety status indicating a completion of the safety condition action and a failure status indicating the detecting of the catastrophic event, and
based on generating the trigger status, copying at least a portion of the contents of the volatile memory to the non-volatile memory.
2. The method as claimed in claim 1 wherein detecting the catastrophic event includes detecting a power error at a power management unit in the hybrid memory unit.
3. The method as claimed in claim 1 wherein initiating the safety condition action includes initiating a self-refresh mode in the volatile memory.
4. The method as claimed in claim 1 wherein detecting the catastrophic event includes detecting a non-operational mode of the external host computing system.
5. The method as claimed in claim 1 wherein initiating the safety condition action includes electrically isolating the hybrid memory unit from the external host computing system.
6. A method of operation of a hybrid memory unit comprising a volatile memory and a non-volatile memory, the method comprising:
detecting a power loss at an external host computing system operably connected to the hybrid memory unit;
upon detecting the power loss, receiving power from an alternate power source;
initiating a self-refresh mode for the volatile memory of the hybrid memory unit;
generating a safety status indicating the volatile memory is in self-refresh mode;
generating a trigger status by detecting both a successful completion of a safety condition action and a failure status from the external host computing system, wherein the safety condition action is initiated based on receiving the failure status at a registering clock driver of the hybrid memory unit;
in response to generating the trigger status, copying at least a portion of the contents of the volatile memory to the non-volatile memory; and
restoring the at least the portion of the contents copied to the non-volatile memory to the volatile memory.
7. The method as claimed in claim 6 wherein detecting the catastrophic event includes receiving a failure flag from the external host computing system on a status pin.
8. The method as claimed in claim 6 wherein detecting the catastrophic event includes detecting a failure of a primary power line in the hybrid memory unit.
9. The method as claimed in claim 6 further comprising:
receiving, by the hybrid memory unit, a primary power line input; and
generating a volatile memory operating voltage and non-volatile memory operating voltage using the primary power line input.
10. A hybrid memory unit, comprising:
a memory board having a memory bus for coupling to an external computing system;
a volatile memory coupled to a non-volatile memory, the volatile memory and the non-volatile memory on the memory board;
a status pin on the memory bus of the memory board, the status pin configured to carry a system status from the external computing system;
a first controller on the memory board for controlling the volatile memory and the non-volatile memory, the first controller configured to detect the system status from the status pin;
a registering clock driver device on the memory board, the registering clock driver device configured to:
determine a volatile memory state;
execute a safety condition action for the volatile memory; and
generate a safety status based on the completion of the safety condition action;
a power management unit on the memory board, the power management unit coupled to a primary power pin and configured to switch to an alternate power source upon detection of a catastrophic event;
a trigger unit coupled to both the registering clock driver device and the first controller, the trigger unit configured to generate a trigger status based on the safety status indicating the completion of the safety condition action and a failure status indicating a backup operation of the volatile memory should be performed; and
a second controller on the memory board, the second controller coupled to the power management unit and the registering clock driver, the second controller configured to:
detect the catastrophic event on the external computing system;
generate the failure status;
receive the trigger status from the trigger unit; and
based on the trigger unit generating the trigger status, initiate copying of at least a portion of the volatile memory to the non-volatile memory.
11. The system as claimed in claim 10 wherein the power management unit is configured to detect a power error at the power management unit.
12. The system as claimed in claim 10 wherein the safety condition action for the volatile memory includes an initiation of a self-refresh mode for the volatile memory.
13. The system as claimed in claim 10 wherein the second controller is configured to detect the catastrophic event by detecting a non-operational mode of the external computing system.
14. The system as claimed in claim 10 wherein the safety condition action for the volatile memory includes electrically isolating the hybrid memory unit from the external computing system.
15. The system as claimed in claim 10 wherein:
the catastrophic event is a power loss;
the safety condition action for the volatile memory is a self-refresh mode for the volatile memory; and
the safety status indicates the volatile memory is in the self-refresh mode.
16. The system as claimed in claim 15 wherein the second controller is configured to detect the catastrophic event by receiving a failure flag from the external computing system on the status pin.
17. The system as claimed in claim 15 wherein the catastrophic event is a failure of a primary power line.
18. The system as claimed in claim 15 wherein the power management unit is configured to:
receive a primary power line input; and
generate a volatile memory operating voltage and non-volatile memory operating voltage using the primary power line input.
19. The system as claimed in claim 15 wherein the registering clock driver is configured to initiate the safety condition action based on receiving the failure status.