Data storage device and method for direct data quantization in multi-level-cell memory
In a data storage device with a multi-level-cell memory, bits of data are stored in different levels of each of the memory cells. In some situations, a quantized version of the stored data (e.g., the most-significant bits) may be requested. Responding to such a request can involve reading all levels of the memory cells to retrieve a full version of the data and then selectively providing only the quantized version. To improve performance, the data is stored in the memory in an interleaved manner in which the quantized version of the data is stored in the same level(s) instead of being spread across all levels of the memory cells. That way, when the quantized version of the data is later requested, only the relevant level(s) are sensed, thereby avoiding the time and resources needed to read memory level(s) that do not store the quantized data.
1 . A data storage device comprising:
a memory comprising multi-level memory cells; and
one or more processors, individually or in combination, configured to:
receive a request to store data;
store the data in the multi-level memory cells, wherein each bit of a set of most-significant bits of the data is stored in a same memory cell level in each respective multi-level memory cell;
in response to receiving a request to read a quantized version of the data, read the set of most-significant bits of the data stored in the same memory cell level of each respective multi-level memory cells without reading other bits of the data stored in at least one other memory cell level of the multi-level memory cells; and
in response to receiving a request to read a non-quantized version of the data, read the other bits of the data stored in the at least one other memory cell level of the multi-level memory cells;
wherein the other bits of the data stored in the at least one other memory cell level of the multi-level memory cells are read when the non-quantized version of the data is read but not when the quantized version of the data is read.
2 . The data storage device of claim 1 , wherein:
the set of most-significant bits of the data is stored in a first memory cell level of each respective multi-level memory cell;
each bit of a set of next-most-significant bits of the data is stored at a second memory cell level in each respective multi-level memory cell; and
the one or more processors, individually or in combination, are further configured to read both the set of most-significant bits of the data and the set of next-most-significant bits of the data in response to receiving the request to read the quantized version of the data.
3 . The data storage device of claim 1 , wherein the one or more processors, individually or in combination, are further configured to:
in response to receiving the request to read the non-quantized version of the data:
read the set of most-significant bits of the data from the same memory cell level in each respective multi-level memory cells; and
read other bits of the data from other memory cell levels of the multi-level memory cells.
4 . The data storage device of claim 1 , wherein:
the multi-level memory cells comprise quad-level cells; and
the set of most-significant bits of the data is stored in a page of the quad-level cells that requires a fewest number of memory senses.
5 . The data storage device of claim 4 , wherein:
a set of next-most-significant bits of the data is stored in an upper page of the quad-level cells.
6 . The data storage device of claim 1 , wherein:
the multi-level memory cells comprise quad-level cells; and
the set of most-significant bits of the data is stored in an upper page of the quad-level cells.
7 . The data storage device of claim 1 , wherein each bit of the set of most-significant bits of the data is stored in the same memory cell level in each respective multi-level memory cell in response to the request specifying a logical block address that is in a designated logical block address range.
8 . The data storage device of claim 1 , wherein each bit of the set of most-significant bits of the data is stored in the same memory cell level in each respective multi-level memory cell in response to the request being received from a designated host.
9 . The data storage device of claim 1 , wherein the data storage device is an artificial intelligence/machine learning (AI/ML) specialized data storage device.
10 . The data storage device of claim 1 , wherein the data storage device is part of a security system.
11 . The data storage device of claim 1 , wherein the memory comprises a three-dimensional memory.
12 . In a data storage device comprising multi-level memory cells, a method comprising:
receiving a request to store data;
storing the data in the multi-level memory cells, wherein each bit of a set of most-significant bits of the data is stored in a same memory cell level in each respective multi-level memory cell;
in response to receiving a request to read a quantized version of the data, reading the set of most-significant bits of the data stored in the same memory cell level of each respective multi-level memory cells without reading other bits of the data stored in at least one other memory cell level of the multi-level memory cells; and
in response to receiving a request to read a non-quantized version of the data, reading the other bits of the data stored in the at least one other memory cell level of the multi-level memory cells;
wherein the other bits of the data stored in the at least one other memory cell level of the multi-level memory cells are read when the non-quantized version of the data is read but not when the quantized version of the data is read.
13 . The method of claim 12 , further comprising:
in response to receiving the request to read the non-quantized version of the data:
sensing all memory cell levels in the multi-level memory cells; and
responding to the request to read the non-quantized version of the data by returning bits sensed from all of the memory cell levels in the multi-level memory cells.
14 . The method of claim 12 , wherein:
the multi-level memory cells comprise quad-level cells; and
the set of most-significant bits of the data is stored in a page of the quad-level cells that requires a fewest number of memory senses.
15 . The method of claim 14 , wherein:
a set of next-most-significant bits of the data is stored in an upper page of the quad-level cells.
16 . The method of claim 12 , wherein:
the multi-level memory cells comprise quad-level cells; and
the set of most-significant bits of the data is stored in an upper page of the quad-level cells.
17 . The method of claim 12 , wherein each bit of the set of most-significant bits of the data is stored in the same memory cell level in each respective multi-level memory cell in response to a write request for the data specifying a logical block address that is in a designated logical block address range.
18 . The method of claim 12 , wherein each bit of the set of most-significant bits of the data is stored in the same memory cell level in each respective multi-level memory cell in response to a write request for the data being received from a designated host.
19 . A data storage device comprising:
a memory comprising multi-level memory cells; and
means for:
receiving a request to store data;
storing the data in the multi-level memory cells, wherein each bit of a set of most-significant bits of the data is stored in a same memory cell level in each respective multi-level memory cell;
in response to receiving a request to read a quantized version of the data, reading the set of most-significant bits of the data stored in the same memory cell level of each respective multi-level memory cells without reading other bits of the data stored in at least one other memory cell level of the multi-level memory cells; and
in response to receiving a request to read a non-quantized version of the data, reading the other bits of the data stored in the at least one other memory cell level of the multi-level memory cells;
wherein the other bits of the data stored in the at least one other memory cell level of the multi-level memory cells are read when the non-quantized version of the data is read but not when the quantized version of the data is read.