Micron Technology Incorporated

Daniel A Boals

Byron D Harris

Karl David Schuh

Abedon Madril

Amy Lee Wohlschlegel

Sanjay Subbarao

Johnny A Lam

Tom V Geukens

Samyukta Mudugal

A total count for an address mapping table is maintained, wherein the total count reflects a total number of updates to the address mapping table, and wherein the address mapping table comprises a plurality of sections. Respective section counts for the plurality of sections are maintained, wherein each respective section count reflects a total number of updates to a corresponding section. It is determined that the total count for the address mapping table satisfies a threshold criterion. A first section of the plurality of sections with a highest section count is identified based on the respective section counts. The first section of the address mapping table is written to a non-volatile memory device.

Managing power loss recovery using a dirty section write policy for an address mapping table in a memory sub-system

A journal count reflecting a number of logical-to-physical (L2P) journals written to a non-volatile memory device is maintained, wherein each L2P journal is associated with one or more updates to an L2P address mapping table. In response to determining that the journal count satisfies a threshold criterion, a first section of a plurality of sections of the L2P address mapping table is identified, wherein the plurality of sections of the L2P address mapping table is cached in a volatile memory device. The first section of the L2P address mapping table is written to the non-volatile memory device.

Managing power loss recovery using an oldest section write policy for an address mapping table in a memory sub-system

A logical-to-physical (L2P) table is maintained, wherein a plurality of sections of the L2P table is cached in a volatile memory device. A total dirty count for the L2P table is maintained, wherein the total dirty count reflects a total number of updates to the L2P table. Respective section dirty counts for the plurality of sections are maintained, wherein each respective section dirty count reflects a total number of updates to a corresponding section. It is determined that the total dirty count for the L2P table satisfies a threshold criterion. In response to determining that the total dirty count for the L2P table satisfies the threshold criterion, a first section of the plurality of sections is identified based on the respective section dirty counts. The first section of the L2P table is written to a non-volatile memory device.

A data item is programmed to a first set of management units (MUs) associated with a first portion of one or more memory devices. The first portion includes memory cells of a first type. The first set of MUs is associated with a first physical address. A mapping is generated in a virtual MU data structure that associates the first physical address with a set of virtual MUs associated with the one or more memory devices. An entry associated with the data item is added to a logical-to-physical (L2P) table associated with the one or more memory devices. The entry includes an identifier associated with the set of virtual MUs associated with the one or more memory devices. A detection is made that the data item is programmed to a second set of MUs associated with a second portion of the one or more memory devices. The second portion includes memory cells of a second type. The second set of MUs is associated with a second physical address. The mapping is updated to associate the set of virtual MUs with the second physical address.

Virtual management unit scheme for two-pass programming in a memory sub-system

A zone is loaded onto a first memory component of a storage system, wherein the zone comprises one or more regions of data blocks comprising a first plurality of logical block addresses (LBAs), and a snapshot of each of the one or more regions is stored on a second memory component of the storage system and is associated with a version identifier. A particular version identifier associated with a respective snapshot of a region is identified, and a set of journals stored on the second memory component are identified, wherein the set of journals comprise a second plurality of LBAs mapped to a second plurality of physical block addresses. The one or more regions of the zone are reconstructed, wherein the reconstructing comprises reading the second plurality of LBAs of the set of journals to identify a first logical block address (LBA) that matches with a second LBA within the first plurality of LBAs on the loaded zone on the first memory component, and updating a physical block address corresponding to the second LBA on the loaded zone with a physical block address corresponding to the first LBA from the set of journals.

Zone based reconstruction of logical to physical address translation map

A logical-to-physical (L2P) address mapping table is maintained, wherein a plurality of sections of the L2P address mapping table is cached in a volatile memory device. A journal entry count is maintained reflecting a number of L2P journal entries associated with an L2P journal. It is determined that the journal entry count satisfies a first threshold criterion. In response to determining that the journal entry count satisfies the first threshold criterion, a writing of the L2P journal to a non-volatile memory device is triggered. A written journal count reflecting a number of L2P journals written to the non-volatile memory device is maintained. In response to determining that the written journal count satisfies a second threshold criterion, a first section of the plurality of sections of the L2P address mapping table is identified. The first section of the L2P address mapping table is written to the non-volatile memory device.

An operation performed on a non-volatile memory device is detected. A type of the operation is determined. The type of the operation comprises a type identifier and a type modifier. A journal entry is generated reflecting the operation. The journal entry comprises a body reflecting the operation and a header comprising the type identifier and the type modifier. A combination of the type identifier and the type modifier define a size of the journal entry. The journal entry is written to a volatile memory device.

Journal scheme for a memory sub-system

A first data location is identified, where the first data location is predetermined to store a first parity data based on a plurality of user data at a plurality of data locations of a storage system. A set of user data is stored at a subset of data locations of the plurality of data locations. A second parity data is stored at a memory buffer location. The second parity data is generated based on the set of user data. An indication of an occurrence of an event is received, the event associated with the storage system. In response to receiving the indication of the occurrence of the event, the second parity data is stored at a second data location external to the plurality of data locations. Upon restart of the storage system, the second parity data is transferred to a third data location of the plurality of data locations. The third data location is situated after the subset of the plurality of data locations and prior to the first data location. The memory buffer is initialized with an initial value of a default hardware initialization state of the storage system to allow for calculating additional parity data based on the initial value and additional user data to be stored in one or more data locations situated after the third data location.

Storing parity data mid stripe

A plurality of generators that each correspond to a respective one of a plurality of cursors associated with the plurality of memory devices are identified. A sequence of logical unit numbers (LUNs) of a plurality of sequences of LUNs are generated for each of the cursors based on a respective generator corresponding to the plurality of cursors. Each of the cursors are directed to perform a memory operation on a set of LUNs in an order provided by the sequence of LUNs that is associated with each of the plurality of cursors.

Per cursor logical unit number sequencing

A portion of a logical block address to physical block address ('L2P') translation map may be identified. A last snapshot of the portion of the L2P translation map may be identified. One or more write operations may be determined, where the write operations are associated with logical block addresses of the portion of the L2P translation map. The write operations may have been performed after the last snapshot of the portion of the L2P translation map was stored. An address on the portion of the L2P translation map may be updated by a processing device based on the determined one or more write operations and the last snapshot of the portion of the L2P translation map.

Each of a multiple cursors is assigned a respective generator of multiple generators. The cursors are used to perform an operation on a set of logical unit numbers (LUN) associated with memory devices. Multiple sequences of LUNs are identified based on the generators. Each of the cursors is associated with one of the sequences of LUNs. The operation on the set of LUNs associated with the memory devices is performed using the sequences of LUNs. The operation on the set of LUNs is performed in an order provided by the sequences of LUNs to reduce a probability of collision by the cursors in the performance of the operation.

User data to be stored at a data block location of a plurality of data block locations at a storage system may be received. A parity data based on the received user data may be generated. An indication of an occurrence of an event associated with the storage system may be received. In response to receiving the indication of the occurrence of the event, the parity data may be stored by a processing device at a particular data block location of the plurality of data block locations where the particular data block location is situated prior to another data block location, of the plurality of data block locations, that is to store another parity data based on a subsequent user data.

Growing community of inventors

Broomfield, CO, United States of America

Daniel A Boals