Wear leveling for non-volatile memory using data write counters
A memory system has a controller (e.g., CPU, FPGA, or GPU) and recording segments in a non-volatile memory (e.g., a flash memory device) used by the controller to store data. The controller is configured to: maintain data write counters for the recording segments; select a first segment of the recording segments for recording data from a host system, wherein selecting the first segment comprises scanning the data write counters to identify a first data write counter corresponding to the first segment; receive, from the host system, data to be recorded by the non-volatile memory; and write the received data to the selected first segment.
1. A method comprising:
receiving, from at least one sensor associated with a navigation system of a vehicle, sensor data;
controlling the vehicle using a neural network having the sensor data as input;
determining, based on the sensor data, a size for first and second memory segments of a non-volatile memory device;
receiving, by the memory device, first data from a host;
scanning, by a controller of the memory device, a first and a second write counter of the memory device for determining in which memory segment to store the first data, the first write counter configured to record a number of write operations performed in the first memory segment of the memory device, and the second write counter configured to record a number of write operations performed in the second memory segment of the memory device;
comparing values stored in the first and the second write counter;
performing processing by the neural network to provide result data;
selecting one of the first or second memory segment based on the sensor data and the value comparison, wherein the result data from the neural network is used to select the first or second memory segment; and
writing the first data to the selected memory segment.
2. The method of claim 1 , wherein the host is a black box recorder of a vehicle.
3. The method of claim 1 , wherein the memory device is a flash memory.
4. The method of claim 1 , wherein when the data is to be written to the first memory segment, the data streams into the first memory segment, and wherein when the first memory segment is full, later streamed-in data replaces previously-recorded data in the first memory segment.
5. The method of claim 1 , further comprising incrementing the value of the first write counter after the data is written to the first memory segment.
6. The method of claim 1 , wherein writing the first data to the selected memory segment comprises writing the first data to the first memory segment if the value stored in the first write counter is lower than the value stored in the second write counter.
7. The method of claim 1 , wherein writing the first data to the selected memory segment comprises writing the first data to the second memory segment if the value stored in the second write counter is lower than the value stored in the first write counter.
8. A memory device comprising:
a navigation system of a vehicle;
at least one sensor associated with the navigation system;
an array of memory cells divided into at least a first and a second memory segment;
a first write counter configured to store a first value corresponding to a number of write operations performed in the first memory segment;
a second write counter configured to store a second value corresponding to a number of write operations performed in the second memory segment; and
a controller configured to:
receive, from the at least one sensor, sensor data;
control the vehicle using a neural network having the sensor data as input;
determine, based on the sensor data, a size for the first and second memory segments;
scan the first and the second write counter to obtain the first and the second value upon receiving first data by the memory device for determining in which memory segment to store the data;
compare values stored in the first and the second write counter;
perform processing by the neural network to provide result data;
select one of the first or second memory segment based on the sensor data and the value comparison, wherein the result data from the neural network is used to select the first or second memory segment; and
write the first data to the selected memory segment.
9. The memory device of claim 8 , wherein the memory device is a flash memory.
10. The memory device of claim 8 , wherein the first and the second memory segment are of equal size.
11. The memory device of claim 8 , wherein a predetermined identification indicates a type of memory cells in each of the first and the second memory segment.
12. The memory device of claim 8 , wherein when the data is to be written to the first memory segment, the data streams into the first memory segment, and wherein when the first memory segment is full, later streamed-in data replaces previously-recorded data in the first memory segment.
13. The memory device of claim 8 , wherein the first write counter increments a value stored therein after the data is written to the first memory segment.
14. The memory device of claim 8 , wherein writing the first data to the selected memory segment comprises writing the first data to the first memory segment if the value stored in the first write counter is lower than the value stored in the second write counter.
15. The memory device of claim 8 , wherein writing the first data to the selected memory segment comprises writing the first data to the second memory segment if the value stored in the second write counter is lower than the value stored in the first write counter.
16. The memory device of claim 8 , further comprising a neural network configured to control the vehicle using the sensor data as input.
17. The memory device of claim 16 , wherein a result from the neural network is used to select first or second memory segment.
18. A memory device comprising:
a navigation system of a vehicle configured to operate in an autonomous navigation mode;
at least one sensor associated with the navigation system;
first and second memory segments of a flash memory;
a first write counter configured to store a first value corresponding to a number of write operations performed in the first memory segment;
a second write counter configured to store a second value corresponding to a number of write operations performed in the second memory segment; and
a controller configured to:
receive, from the at least one sensor, sensor data;
control the vehicle using a neural network having the sensor data as input;
determine, based on the sensor data, a size for the first and second memory segments;
receive a command to stop recording;
perform processing by the neural network to provide result data;
in response to receiving the command to stop recording, select either the first or second memory segment based on the sensor data and scanning the first and the second write counters and comparing the first and the second values, wherein the result data from the neural network is used to select the first or second memory segment;
enter into the autonomous navigation mode; and
when the vehicle enters into the autonomous navigation mode, start recording data to the selected first or second memory segment.