Efficient reduction of read disturb errors in NAND flash memory

A method of inhibiting read disturb errors in a non-volatile memory in which memory locations are divided into pages and blocks, with each page corresponding to a specific physical memory location in which a group of bits can be stored, and each block corresponding to a group of pages, the method comprising the steps of:

• maintaining a count of the number of times a READ operation has been performed on any page within a given block using a first predefined increment if no error is detected in any of the pages in the given block after the READ operation and using a second predefined increment if an error is detected in any of the pages in the given block after the READ operation, the second predefined increment being larger than the first predefined increment;
• determining that the count has reached a first predetermined threshold;
• and
• after the count has reached the first predetermined threshold, moving data stored within the given block to memory locations within a different block on a page-by-page basis in response to, and as a result of, READ operations being subsequently performed on pages within the given block, such that each of a plurality of READ operations received after the count has reached the first predetermined threshold results in movement of only a single page of data from the given block.

A memory system comprising:

• a memory space, the memory space containing physical address locations to which data may be stored, the locations being grouped into erase groups, where each erase group comprises a collection of physical address locations that may be erased through a single erase operation;
• and
• means for maintaining a count of the number of times any of the locations within a given erase group have been subjected to a read operation since a defined event using a first predefined increment if no error is detected in any of the locations in the given erase group after the read operation and using a second predefined increment if an error is detected in any of the locations in the given erase group after the read operation, the second predefined increment being larger than the first predefined increment, the means for maintaining including means for:
• (a) determining whether the maintained count has reached a first predetermined threshold and (b) determining whether the maintained count has reached a second predetermined threshold, where the second predetermined threshold is greater than the first predetermined threshold;
• and
• means for responding to READ requests, each READ request being directed to data stored at a logical address that corresponds to a physical address within the given erase group, by:
• (i) if the count is below the first predetermined threshold:
• returning the data stored at the physical address corresponding to the logical address and maintaining as valid the data stored within the corresponding physical address;
• (ii) if the count is at or above the first predetermined threshold but below the second predetermined threshold:
• (a) returning the data stored at the physical address corresponding to the logical address, (b) moving the data stored at the corresponding physical address to a new physical address in a different erase group, (c) maintaining the data stored in the physical address locations of the erase group other than the physical address corresponding to the logical address in the physical address locations where such data was stored at the time the READ request was received and (d) designating the data stored within the corresponding physical address as invalid while maintaining as valid data within the given erase group that was valid at the time the READ request was received;
• and
• (iii) if the count is above the second predetermined threshold:
• (a) returning the data stored at the physical address corresponding to the logical address and (b) moving all data stored within the first erase group that was valid at the time the READ request was received to a different erase group, and (c) designating the data within the physical addresses in the first erase group from which the data was moved as invalid.