Storage device with erase unit level address mapping
This disclosure provides for improvements in managing multi-drive, multi-die or multi-plane NAND flash memory. In one embodiment, the host directly assigns physical addresses and performs logical-to-physical address translation in a manner that reduces or eliminates the need for a memory controller to handle these functions, and initiates functions such as wear leveling in a manner that avoids competition with host data accesses. A memory controller optionally educates the host on array composition, capabilities and addressing restrictions. Host software can therefore interleave write and read requests across dies in a manner unencumbered by memory controller address translation. For multi-plane designs, the host writes related data in a manner consistent with multi-plane device addressing limitations. The host is therefore able to “plan ahead” in a manner supporting host issuance of true multi-plane read commands.
1 . A storage drive comprising:
a host interface;
one or more dies of flash memory, including a first die having planes;
storage subdivisions, each corresponding to a respective set of erase units, the erase units of each set being mutually-exclusive to the erase units of each other set, the set respective to a given one of the storage subdivisions including erase units in respective ones of the planes; and
logic, comprising at least one of circuitry or instructions operable to control the operation of circuitry, the logic structured to:
receive write requests from a host device, via the host interface, wherein the write requests are addressed to the given one of the subdivisions;
store write data, associated with respective ones of the write requests, in the erase units in the respective ones of the planes, according to a common page offset into the respective erase units; and
cause the storage drive to transmit, to the host device, information representing a size and logical address information for the given one of the storage subdivisions.
2 . The storage drive of claim 1 , wherein the logic is further structured to, with circuitry:
receive one or more further commands from the host device, wherein the one or more further commands are addressed to the given one of the storage subdivisions; and
responsive to the one or more further commands, control physical erasure of each erase unit in the set which is respective to the given one of the storage subdivisions, while not, responsive to the one or more further commands, controlling physical erasure of erase units in the sets respective to other ones of the storage subdivisions.
3 . The storage device of claim 1 , wherein the storage device is to receive one or more requests from the host device to deallocate storage of the flash memory associated with the given one of the storage subdivisions, and wherein the logic is further structured to, with circuitry:
detect a condition in which all written storage, associated with the given one of the storage subdivisions, has been deallocated; and
responsive to detection of the condition, automatically control physical erasure of each of erase unit, in the set respective to the given one of the storage subdivisions, which corresponds to the written storage.
4 . The storage device of claim 1 , wherein one or more of the write requests are logically directed to overwrite data written stored in the given one of the storage subdivisions, and wherein:
the logic is further structured to, with circuitry, detect a condition in which all written storage, associated with the given one of the storage subdivisions, has been deallocated; and
responsive to detection of the condition, automatically control physical erasure of each erase unit in the set respective to the given one of the storage subdivisions, which corresponds to the written storage.
5 . The storage drive of claim 1 , wherein the logic is further structured to, with circuitry:
store metadata respective to the given one of the storage subdivisions;
receive a query from the host device; and
responsive to the query, transmit, to the host device, additional information, dependent on the stored metadata, wherein the additional information is specific to the given one of the storage subdivisions.
6 . The storage drive of claim 5 , wherein the additional information is transmitted to the host device via the interface, and indicates an extent to which the given one of the storage subdivisions is full.
7 . The storage drive of claim 5 , wherein the additional information represents a time since data was written to at least one erase unit in the set of erase units which is respective to the given one of the storage subdivisions.
8 . The storage drive of claim 1 , wherein the size of the given one of the storage subdivisions corresponds to a minimum amount of storage capacity to be erased by the storage drive.
9 . The storage drive of claim 1 , wherein the logic is further operable to, with circuitry:
store metadata respective to the given one of the storage subdivisions;
compare the metadata to at least one criterion;
detect a condition where the metadata satisfies the at least one criterion; and
responsive to detection of the condition, copy data from one or more erase units in the set which is respective to the given one of the storage subdivisions to a different erase unit, wherein the different erase unit is not mapped to any of the storage subdivisions other than the given one of the storage subdivisions.
10 . The storage drive of claim 1 , wherein the logic is further structured to, with circuitry, store write data in respective ones of the planes in a manner that overlaps in time.
11 . The storage drive of claim 1 , wherein the logic is further operable to, with circuitry, detect an error condition associated with one of the erase units of the respective set of erase units, and to responsively include a different erase unit within the respective set of erase units.
12 . The storage drive of claim 1 , wherein the logic is further structured to, with circuitry:
track release status information for individual logical block addresses (LBAs) associated with the given one of the storage subdivisions; and
perform at least one maintenance operation on one or more erase units in the set which is respective to the given one of the of the storage subdivisions, to the exclusion of erase units which are not in the set which is respective to the given one of the storage subdivisions.
13 . The storage drive of claim 1 , wherein sequential logical addresses respectively accompany the write requests, and wherein the logic is further structured to, with circuitry, access the erase units in the respective ones of the planes according to the sequential logical addresses.
14 . The storage device of claim 1 , wherein:
the one or more dies of flash memory include a second one of the dies and a third one of the dies; and
the logic is further structured to cause the storage device to, with circuitry:
receive a first write request and a second write request, each of the first write request and the second write requested directed to the given one of the storage subdivisions; and
responsively store data corresponding to the second write request in an erase unit in the second one of the dies and data corresponding to the third write request in the third one of the dies, each according to a common page offset.
15 . The storage drive of claim 1 , wherein the logic is further structured to, with circuitry, receive a single one of the write requests, and responsively control programming of write data into at least one of (a) multiple planes or (b) multiple dies of the flash memory in a manner that overlaps in time.
16 . The storage drive of claim 1 , wherein the logic is further structured to, with circuitry, transmit, to the host device, information representing a quantity of the storage subdivisions.
17 . The storage drive of claim 1 , wherein the logic is further structured to, with circuitry, store metadata, the metadata representing one of wear or erase count, in a manner that is respective to at least one of (a) the given one of the storage subdivisions or (b) erase units in the set which is respective to the given one of the storage subdivisions.
18 . A method of operating a storage drive having a host interface, one or more dies of flash memory, including a first die having planes, and logic, comprising at least one of circuitry or instructions operable to control the operation of circuitry, the method comprising:
establishing storage subdivisions, each corresponding to a respective set of erase units, the erase units of each set being mutually-exclusive to the erase units of each other set, the set respective to a given one of the storage subdivisions including erase units in respective ones of the planes; and
with the logic:
receiving write requests from a host device, via the host interface, wherein the write requests are addressed to the given one of the subdivisions;
storing write data, associated with respective ones of the write requests, in the erase units in the respective ones of the planes, according to a common page offset into the respective erase units; and
transmitting, to the host device, information representing a size and logical address information for the given one of the storage subdivisions.
19 . The method of claim 18 , wherein the method further comprises:
storing metadata respective to the given one of the storage subdivisions;
receiving a query from the host device; and
responsive to the query, transmitting, to the host device, additional information, dependent on the stored metadata, wherein the additional information is specific to the given one of the storage subdivisions.
20 . The method of claim 18 , wherein the method further comprises transmitting to the host device information which indicates an extent to which the given one of the storage subdivisions is full.