Hybrid wear leveling for in-situ data replacement media

By adopting hybrid loss equalization technology in the memory subsystem, using data unit levels and mapping frequency of different sizes, the problem of low loss equalization efficiency of memory components in the prior art is solved, and more efficient resource utilization and memory life extension are achieved. .

CN119046183APending Publication Date: 2024-11-29MICRON TECHNOLOGY INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411157414.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2018-08-23
Filing Date
2019-08-21
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

When existing memory subsystems deal with different types of memory components, it is difficult to achieve efficient loss equalization, resulting in premature failure of specific memory components, and loss equalization operations consume a lot of computing resources and time.

Method used

Using hybrid loss equalization technology, using different sizes of data unit hierarchy, direct algebraic mapping and indirect full correlation mapping are used at the first and second levels respectively, and through the loss equalization operation of different frequencies, uniform distribution of memory components is achieved. and resource optimization.

Benefits of technology

It improves the efficiency and loss balance of the memory subsystem, reduces access delay, extends the service life of memory components, avoids premature failure of specific components, and reduces write amplification overhead and memory footprint.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119046183A_ABST
    Figure CN119046183A_ABST
Patent Text Reader

Abstract

The invention relates to hybrid wear leveling for in-situ data replacement media. A memory subsystem is provided that periodically performs a first wear leveling operation using a direct mapping function at a first frequency on a data management unit of a memory component of the memory subsystem. The memory subsystem further periodically performs a second wear leveling operation using indirect mapping at a second frequency on a group of data management units of the memory component, wherein the second wear leveling operation is performed less frequently than the first wear leveling operation.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Information about divisional applications

[0002] This application is a divisional application of the invention patent application with application date of August 21, 2019 (priority date of August 23, 2018), application number 201980059927.3, and invention name “Hybrid Wear Leveling for In-Place Data Replacement Media”. Technical Field

[0003] Embodiments of the present disclosure relate generally to memory subsystems, and more particularly, to hybrid wear leveling for in-place data replacement media. Background Art

[0004] The memory subsystem may be a storage system, such as a solid-state drive (SSD), and may include one or more memory components that store data. The memory components may be, for example, non-volatile memory components and volatile memory components. In general, the host system may utilize the memory subsystem to store data at the memory components and retrieve data from the memory components. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] The present disclosure will be more fully understood from the detailed description given below and the accompanying drawings of various embodiments of the present disclosure. However, the drawings should not be considered to limit the present disclosure to specific embodiments, but are only for explanation and understanding.

[0006] Figure 1 An example computing environment including a memory system according to some embodiments of the present disclosure is described.

[0007] Figure 2A Different levels of hierarchical data units in a memory component are described according to some embodiments of the present disclosure.

[0008] Figure 2B In-SMU wear leveling to be performed using direct algebraic mapping according to some embodiments of the present disclosure is described.

[0009] Figure 2C Inter-SMU wear leveling to be performed using an indirect map according to some embodiments of the present disclosure is described.

[0010] Figure 3 is a flow chart of an example method of performing hybrid wear leveling for in-place data replacement media according to some embodiments of the present disclosure.

[0011] Figure 4 is a flow chart of an example method of performing hybrid wear leveling for in-place data replacement media according to some embodiments of the present disclosure.

[0012] Figure 5is a block diagram of an example machine of a computer system upon which embodiments of the present disclosure may operate. DETAILED DESCRIPTION

[0013] Aspects of the present disclosure are directed to hybrid wear leveling for in-place data replacement media in a memory subsystem. A memory subsystem is also referred to hereinafter as a "memory device". An example of a memory subsystem is a storage device system, such as a solid-state drive (SSD). In some embodiments, the memory subsystem is a hybrid memory / storage device subsystem. In general, a host system may utilize a memory subsystem that includes one or more memory components. The host system may provide data for storage at the memory subsystem and may request retrieval of data from the memory subsystem.

[0014] Memory components used by a memory subsystem may have specific properties that provide challenges in the operation of the memory subsystem. For example, some memory components, such as non-volatile memory components, have limited endurance. Individual segments, data units, or blocks of a memory component can only be written, read, and / or erased a limited number of times before physical wear and tear causes the memory component to fail. To counteract this limited endurance of memory components, techniques for managing wear on memory components have been developed.

[0015] One technique for managing durability in conventional memory subsystems is wear leveling. Wear leveling operations may attempt to evenly distribute read, write, and erase operations across memory components, and thus corresponding physical wear. A memory component may be divided into a certain number of individual data units, such as pages or blocks of a memory component, each of which experiences physical wear. These data units may represent individual fragments of a memory component that may be written or erased in a single operation. Write counts (e.g., the number of write operations performed on a data unit to write data to a data unit during the useful life of the data unit), read counts (e.g., the number of read operations performed on a data unit to read data from a data unit during the useful life of the data unit), or erase counts (e.g., the number of erase operations performed on a data unit to erase data from a data unit during the useful life of the data unit) may be strong indicators of wear on data units of a memory component. Therefore, conventional wear leveling techniques typically use a sorting procedure to find data units with high read or write counts and data units with low read counts or write counts. Data from data cells having high read or write counts may be swapped with data from data cells having low read or write counts in an attempt to evenly distribute wear across the data cells of the memory component.

[0016] However, different types of memory components may include different numbers of data cells. For example, a cross-point array of non-volatile memory cells may have a significantly smaller data cell size than a flash-based memory component, and therefore may have a significantly larger number of data cells for a memory component of similar capacity. The cross-point array of non-volatile memory cells may be combined with a stackable cross-grid data access array to perform bit storage based on changes in body resistance. In addition, in contrast to many flash-based memories, cross-point non-volatile memories may perform in-place write operations, in which non-volatile memory cells may be programmed without pre-erasing the non-volatile memory cells. Therefore, such memory components may perform "in-place" data replacement. Due to the large number of data cells in this memory component, a large amount of computing resources (e.g., processing and memory resources) are used to retrieve, store, and sort various count values ​​associated with the memory component. In addition, retrieving, storing, and sorting a large number of data cells takes a proportionally large amount of time, which contributes significantly to the latency of the memory subsystem.

[0017] Aspects of the present disclosure address the above and other deficiencies by implementing hybrid wear leveling for in-place data replacement media. In one embodiment, hybrid wear leveling utilizes a hierarchy of data units with different sizes. For example, a first level may represent individual data units, such as data blocks or pages of a memory component (which may also be referred to herein as "management units"), while a second level may represent groups of multiple data units (which may also be referred to herein as "super management units"). Different wear leveling techniques may be used at each level of the hierarchy and wear leveling may be performed at different frequencies. For example, hybrid wear leveling may include wear leveling every five minutes or every 1000 host writes using direct algebraic mapping at the first level and wear leveling every hour or every 10,000 host writes using indirect fully associative mapping implemented by a lookup table at the second level. In other embodiments, the wear leveling techniques and / or associated frequencies may be changed as needed. The hybrid wear leveling scheme described herein implements efficient and effective wear leveling in a memory subsystem with a higher number of data units, such as when in-place data replacement media is used. Processing and memory resources used for wear leveling and data access latency are reduced, and wear on multiple memory components of a memory subsystem (e.g., the number of operations performed on a memory component and / or the amount of data written to a memory component) can be more evenly distributed, thereby preventing premature failure of a particular memory component of the memory subsystem relative to other memory components of the memory subsystem. In addition, there is less write amplification overhead and the memory footprint used for wear leveling is reduced. Figures 1 to 5 Provides additional details on hybrid wear leveling.

[0018] Figure 1An example computing environment 100 including a memory subsystem 110 according to some embodiments of the present disclosure is described. The memory subsystem 110 may include media, such as memory components 112A to 112N. The memory components 112A to 112N may be volatile memory components, non-volatile memory components, or a combination of such components. In some embodiments, the memory subsystem is a storage system. An example of a storage system is an SSD. In some embodiments, the memory subsystem 110 is a hybrid memory / storage device subsystem. In general, the computing environment 100 may include a host system 120 that uses the memory subsystem 110. For example, the host system 120 may write data to the memory subsystem 110 and read data from the memory subsystem 110.

[0019] The host system 120 may be a computing device, such as a desktop computer, a laptop computer, a network server, a mobile device, or such computing devices including a memory and a processing device. The host system 120 may include or be coupled to the memory subsystem 110, so that the host system 120 can read data from the memory subsystem 110 or write data to the memory subsystem 110. The host system 120 may be coupled to the memory subsystem 110 via a physical host interface. As used herein, "coupled to" generally refers to a connection between components, which may be an indirect communication connection or a direct communication connection (e.g., without an intermediate component), whether wired or wireless, including connections such as electrical connections, optical connections, magnetic connections, etc. Examples of physical host interfaces include, but are not limited to, a serial advanced technology attachment (SATA) interface, a peripheral component interconnect express (PCIe) interface, a universal serial bus (USB) interface, a fiber channel, a serial attached SCSI (SAS), etc. The physical host interface may be used to transmit data between the host system 120 and the memory subsystem 110. When the memory subsystem 110 is coupled to the host system 120 through a PCIe interface, the host system 120 can further utilize an NVM Express (NVMe) interface to access the memory components 112A to 112N. The physical host interface can provide an interface for passing control, address, data, and other signals between the memory subsystem 110 and the host system 120.

[0020] Memory components 112A to 112N may include any combination of different types of non-volatile memory components and / or volatile memory components. Examples of non-volatile memory components include non-AND (NAND) type flash memory. Each of memory components 112A to 112N may include one or more arrays of memory cells, such as single-level cells (SLC) or multi-level cells (MLC) (e.g., triple-level cells (TLC) or quad-level cells (QLC)). In some embodiments, a particular memory component may include both an SLC portion and an MLC portion of a memory cell. Each of the memory cells may store one or more bits of data (e.g., a data block) used by the host system 120. Although non-volatile memory components such as NAND type flash memory are described, memory components 112A to 112N may be based on various other types of memory such as volatile memory. In some embodiments, the memory components 112A to 112N may be, but are not limited to, random access memory (RAM), read-only memory (ROM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), phase change memory (PCM), magnetic random access memory (MRAM), or non-(NOR) flash memory, electrically erasable programmable read-only memory (EEPROM), and a cross-point array of non-volatile memory cells. The cross-point array of non-volatile memory may be combined with a stackable cross-grid data access array to perform bit storage based on changes in body resistance. In addition, in contrast to many flash-based memories, cross-point non-volatile memories may perform write-in-place operations, where non-volatile memory cells may be programmed without pre-erasing the non-volatile memory cells. Furthermore, the memory cells of the memory components 112A to 112N may be grouped into memory pages or data blocks, which may refer to cells of a memory component used to store data.

[0021] The memory system controller 115 (hereinafter referred to as the "controller") can communicate with the memory components 112A to 112N to perform operations, such as reading data, writing data, or erasing data at the memory components 112A to 112N, and other such operations. The controller 115 may include hardware, such as one or more integrated circuits and / or discrete components, buffer memory, or a combination thereof. The controller 115 may be a microcontroller, a dedicated logic circuit (e.g., a field programmable gate array (FPGA), an application-specific integrated circuit (ASIC), etc.), or other suitable processor. The controller 115 may include a processor (processing device) 117 configured to execute instructions stored in a local memory 119. In the illustrated example, the local memory 119 of the controller 115 includes an embedded memory configured to store instructions for executing various processes, operations, logic flows, and routines that control the operation of the memory subsystem 110, including manipulating the communication between the memory subsystem 110 and the host system 120. In some embodiments, the local memory 119 may include memory registers that store memory pointers, fetched data, etc. Local memory 119 may also include read-only memory (ROM) for storing microcode. Figure 1 The example memory subsystem 110 in FIG. 1 has been illustrated as including a controller 115, but in another embodiment of the present disclosure, the memory subsystem 110 may not include the controller 115, but may rely on external control (e.g., provided by an external host or by a processor or controller separate from the memory subsystem).

[0022] In general, the controller 115 may receive commands or operations from the host system 120 and may convert the commands or operations into instructions or appropriate commands to achieve the desired access to the memory components 112A to 112N. The controller 115 may be responsible for other operations such as wear leveling operations, garbage collection operations, error detection and error correction code (ECC) operations, encryption operations, cache operations, and address translation between logical block addresses and physical block addresses associated with the memory components 112A to 112N. The controller 115 may further include a host interface circuit to communicate with the host system 120 via a physical host interface. The host interface circuit may convert commands received from the host system into command instructions to access the memory components 112A to 112N, and convert responses associated with the memory components 112A to 112N into information for the host system 120.

[0023] The memory subsystem 110 may also include additional circuits or components not illustrated. In some embodiments, the memory subsystem 110 may include a cache or buffer (e.g., DRAM) and address circuits (e.g., row decoders and column decoders) that may receive addresses from the controller 115 and decode the addresses to access the memory components 112A to 112N.

[0024] The memory subsystem 110 includes a hybrid wear leveling management component 113 that can be used to implement hybrid wear leveling across the memory components 112A to 112N in the memory subsystem 110. In some embodiments, the controller 115 includes at least a portion of the hybrid wear leveling management component 113. For example, the controller 115 may include a processor 117 (processing device) that is configured to execute instructions stored in a local memory 119 for performing the operations described herein. In some embodiments, the hybrid wear leveling management component 113 is part of a host system 120, an application, or an operating system.

[0025] The hybrid wear leveling management component 113 may perform wear leveling using a hierarchy of data units of different sizes. For example, a first level may represent individual data units, such as data blocks or pages of a memory component (i.e., management units), while a second level may represent groups of multiple data units (i.e., super management units). Wear leveling may be performed using different wear leveling techniques at each level of the hierarchy and at different frequencies. For example, the hybrid wear leveling management component 113 may implement direct algebraic mapping at a first frequency at the first level, and may implement indirect fully associative mapping implemented by a lookup table at a second frequency at the second level. In one embodiment, since the management units in the first level are smaller in size than the super management units in the second level, wear leveling is performed more frequently at the first level than at the second level. Additional details regarding the operation of the hybrid wear leveling management component 113 are described below.

[0026] In an embodiment, wear leveling operations may include operations that extend the useful life of memory components 112A-112N (generally referred to herein as "memory components 112"). For example, wear leveling operations may attempt to evenly distribute physical wear across a collection of data units of memory component 112. A data unit may refer to an amount of physical memory of memory component 112.

[0027] Figure 2AThe different levels of hierarchical data units in the memory component 112 are illustrated. A management unit (MU) is one example of a data unit. For example, a management unit may be a page (e.g., the smallest data unit that can be written to or read from the cross-point array) of a non-volatile memory or a block (e.g., the smallest data unit that can be erased in a flash-based memory) of a flash-based memory. A super management unit (SMU) is another example of a data unit and may include a group of multiple management units (e.g., 1000 MUs). The group of data units may be the data units that make up one or more memory components 112. For example, the group of data units of a memory component 112 may be all or most of a page or block of a memory component 112. Figure 2A An example of two super management units 210 and 220 is illustrated. Super management unit 210 includes multiple management units 212, 214, 216, and 218. Super management unit 220 also includes multiple management units 222, 224, 226, and 228. Only a limited number of management units are illustrated, but in practice there may be any number of super management units, each containing any number of management units (e.g., 500, 1000, 2000).

[0028] In one embodiment, the hybrid wear leveling management component 113 may perform wear leveling using different wear leveling techniques and at different frequencies at each level of the hierarchy. For example, the hybrid wear leveling management component 113 may perform intra-SMU wear leveling at a first frequency using direct algebraic mapping among the management units 212, 214, 216, and 218 of the super management unit 210 or the management units 222, 224, 226, and 228 of the super management unit 220. Furthermore, the hybrid wear leveling management component 113 may perform inter-SMU wear leveling at a second, less frequent frequency using indirect fully associative mapping implemented by a lookup table among the super management units 210 and 220 themselves.

[0029] In some implementations, the wear leveling operation may rewrite data of a data unit having a high wear metric to another data unit having a lower wear metric, or vice versa (e.g., rewrite data of a data unit having a low wear metric to another data unit having a higher wear metric). In an implementation, the wear metric may indicate a level of physical wear on the data unit. Some examples of the wear metric may include a write count, a read count, or a combination of a write count and a read count.

[0030] In some implementations, the wear metric may include a combination of a write count and a read count. For example, the wear metric may include both a write count and a read count for a particular data unit, where the write count and the read count are combined using one or more equations. The physical wear on the data unit caused by a read operation may be less than the physical wear caused by a write operation. To combine the read count and the write count of a data unit, the read count or the write count of the data unit may be normalized by weighting the write count or the read count (e.g., adjusting the counts measured by different scales to a common scale). For example, the read count of the data unit may be multiplied by a fraction (e.g., 0.8) to determine the equivalent number of write counts that the read count represents in terms of wear. The weighted read count and write count of the data unit may be combined (e.g., added) and used as a wear metric indicating physical wear on a particular data unit.

[0031] In some embodiments, the memory component 112 may include a non-volatile memory device that includes a cross-point array of non-volatile memory cells. As mentioned above, the cross-point array of the non-volatile memory can be combined with a stackable cross-grid data access array to perform bit storage based on changes in body resistance. In addition, compared to many flash-based memories that perform off-site write operations (e.g., erasing data at a location before other data can be programmed to the location to be written), the cross-point non-volatile memory can perform in-place write operations, where the non-volatile memory cells can be programmed without pre-erasing the non-volatile memory cells. It can be noted that due to the ability of the cross-point non-volatile memory to perform in-place write operations, the wear metric of the cross-point non-volatile memory may not include an erase count. The erase count may be a value indicating the number of memory operations (e.g., erase operations that erase data of the data cell) performed on the data cell during the useful life of the data cell. Aspects of the present disclosure may be applicable to other types of non-volatile memory devices or other types of memory devices.

[0032] Figure 2BIntra-SMU wear leveling to be performed using direct algebraic mapping according to some embodiments of the present disclosure is illustrated. In one embodiment, an algebraic function may be used to determine an appropriate physical location to which data may be moved in a wear leveling operation (e.g., Fn(logical MU index)=physical MU index). Depending on the embodiment, the direct mapping function may include at least one of a swap function, a circular shift function, a linear function utilizing a base pointer value and an idle pointer value, or some other appropriate function. Because the function results in a direct mapping of logical indexes to new physical locations, this intra-SMU wear leveling may be performed more frequently and with lower resource utilization at the first level of the hierarchy than at higher levels. In one embodiment, the hybrid wear leveling management component 113 maintains a base pointer value 231 and an idle pointer value 235 for a super management unit 230 including management units 232, 234, 236, and 238. The base pointer value 231 indicates the location of the starting point of the super management unit 230 (i.e., the location of the first management unit 232). The free pointer value 235 indicates the location of the available space for data storage within the super management unit 230 (i.e., the location of the management unit 236). In one embodiment, the hybrid wear leveling management component 113 increments the free pointer value 235 after every N_free writes, where N_free can be a fixed number, a random number, or a pseudo-random number. The hybrid wear leveling management component 113 further decrements the base pointer value 231 after every N_base steps, where N_base represents the size of the super management unit 230. In one embodiment, the direct mapping function can be expressed as:

[0033] if(Logical MU Index+Base Pointer)mod SMU Size>Free Pointer,

[0034] Physical MU Index=(Logical MU Index+Base Pointer)mod SMU Size;

[0035] else,

[0036] Physical MU Index=(Logical MU Index+Base Pointer)mod SMU Size+1

[0037] Figure 2CInter-SMU wear leveling to be performed using indirect mapping according to some embodiments of the present disclosure is described. Using indirect mapping, information is added to determine the target physical location on the memory component 112. In one embodiment, the hybrid wear leveling management component 113 copies the data from the super management unit to an available physical location on the memory component 112. This physical location can be determined in any number of ways, such as the location with the lowest write count, the location with the least recent access, etc. A list of available locations can be maintained in a free pool stored in the local memory 119. After copying the data, the hybrid wear leveling management component 113 records the mapping of the logical index associated with the data from the super management unit to the available physical index in the lookup table 240. In one embodiment, the lookup table 240 is maintained in the local memory 119 of the controller 115. Therefore, when the logical index is applied to the lookup table 240, the corresponding physical address is output. In other embodiments, the mapping can be maintained in some other type of data structure, such as an array, a linked list, etc. Because indirect mapping utilizes local memory 119 and has associated access latency, this wear leveling operation may not be performed as frequently at the second level of the hierarchy as at lower levels.

[0038] Figure 3 is a flow chart of an example method 300 for performing hybrid wear leveling for in-place data replacement media according to some embodiments of the present disclosure. The method 300 may be performed by processing logic, which may include hardware (e.g., a processing device, a circuit, dedicated logic, programmable logic, microcode, hardware of a device, an integrated circuit, etc.), software (e.g., instructions running or executed on a processing device), or a combination thereof. In some embodiments, the method 300 is performed by Figure 1 The hybrid wear leveling management component 113 of the embodiment of the present invention is executed. Although shown in a specific order or sequence, the order of the processes can be modified unless otherwise specified. Therefore, it should be understood that the illustrated embodiments are only examples, and the illustrated processes can be performed in a different order, and some processes can be performed in parallel. In addition, one or more processes can be omitted in various embodiments. Therefore, not all processes are required in every embodiment. Other process flows are possible.

[0039] At block 310, the processing device detects the occurrence of a first trigger. The first trigger may cause the hybrid wear leveling management component 113 to initiate Figure 2A1. In one embodiment, the first tier may be the lowest tier in the tier hierarchy shown in FIG. 1 and may include the smallest data unit of the memory component 112 that may be controlled by the controller 115. In one embodiment, these data units may include management units, such as data blocks or pages of the memory component 112. In one embodiment, the first trigger is based on at least one of a time period that has passed since a previous wear leveling operation or a number of data write operations performed on the memory component 112 at the request of the host system 120 since the previous wear leveling operation was performed. In one embodiment, the controller 115 includes a timer that is reset each time a wear leveling operation is performed on the first tier. The controller 115 may notify the hybrid wear leveling management component 113 each time the timer expires or reaches a predetermined value. The timer may have a configurable value (e.g., 1 minute, 5 minutes, 10 minutes, etc.) that may be adjusted according to the specific memory subsystem 110. This notification may trigger the hybrid wear leveling management component 113 to initiate a wear leveling operation. In one embodiment, the controller 115 includes a counter that is incremented (or decremented) each time a write request is received from the host system 120 and executed on the memory component 112. The controller 115 may notify the hybrid wear leveling management component 113 each time the counter reaches a predetermined value (e.g., 500 writes, 1000 writes, 2000 writes, etc.). This notification may trigger the hybrid wear leveling management component 113 to initiate a wear leveling operation.

[0040] At block 320, the processing device redistributes the plurality of data units within the first group of data units to different physical locations on the memory component 112A according to a direct mapping function. The direct mapping function may be part of a wear leveling operation and may include any mathematical function, such as an algebraic function, that is complete in itself and does require the use of any additional information in order to determine the target physical location in the memory component 112A. In one embodiment, the hybrid wear leveling management component 113 applies a first logical index associated with data from one of the data units in the first level of the hierarchy to the direct mapping function to determine a physical index corresponding to a physical location on the memory component 112A and then copies the data from the data management unit to the physical location. In one embodiment, the direct mapping function includes at least one of a swap function, a circular shift function, a linear function utilizing a base pointer value and a free pointer value, or some other suitable function. Because the function results in a direct mapping of a logical index to a new physical location, this wear leveling operation may be performed more frequently and with lower resource utilization at the first level of the hierarchy than at higher levels.

[0041] At block 330, the processing device detects the occurrence of a second trigger, wherein the second trigger occurs less frequently than the first trigger. The second trigger may cause the hybrid wear leveling management component 113 to initiate a wear leveling operation of a second level in the hierarchy of levels. In one embodiment, the second level may be a higher level in the hierarchy and may include larger data units of the memory component 112A that may be controlled by the controller 115. In one embodiment, these data units may include super management units, such as groups of data blocks or pages of the memory component 112A. In one embodiment, the second trigger is also based on at least one of a time period that has passed since a previous wear leveling operation or a number of data write operations performed on the memory component 112A at the request of the host system 120 since the previous wear leveling operation was performed. For example, the controller 115 may notify the hybrid wear leveling management component 113 whenever a timer expires or reaches a predetermined value (e.g., 1 hour, 2 hours, 12 hours, 24 hours, etc.) or whenever a write counter reaches a predetermined value (e.g., 5000 writes, 10,000 writes, 20,000 writes, etc.). This notification may trigger the hybrid wear leveling management component 113 to initiate a wear leveling operation.

[0042] At block 340, the processing device redistributes the multiple groups of data units to different physical locations on the memory component 112A using an indirect mapping. With an indirect mapping, information needs to be added in order to target a target physical location on the memory component 112A. In one embodiment, the hybrid wear leveling management component 113 copies the data from the group of data management units to an available physical location on the memory component 112A. This physical location can be determined in any number of ways, such as the location with the lowest write count, the location with the least recent access, and the like. After copying the data, the hybrid wear leveling management component 113 records the mapping of the logical index associated with the data from the group of data management units to the available physical location in a lookup table. In one embodiment, the lookup table is maintained in the local memory 119 of the controller 115. In other embodiments, the mapping may be maintained in some other type of data structure, such as an array, a linked list, and the like. Because the indirect mapping utilizes the local memory 119 and has an associated access latency, this wear leveling operation may not be performed as frequently at the second level of the hierarchy as at the lower levels.

[0043] In one embodiment, the wear leveling operation performed on the second level in the hierarchical hierarchy utilizes both mapped groups of data management units and unmapped groups of data management units. The mapped groups may include physical groups of data management units that are mapped to a specific logical space, while the unmapped groups are not mapped to any specific logical space. Therefore, a given memory component may include at least some mapped groups and some unmapped groups. In one embodiment, the hybrid wear leveling management component 113 determines whether the difference between the highest wear metric and the lowest wear metric on the memory component is greater than a predefined threshold. If so, the hybrid wear leveling management component 113 determines whether the lowest wear metric from the unmapped group is greater than the lowest wear metric from the mapped group. If so, the hybrid wear leveling management component 113 may swap the physical locations of the unmapped group with the lowest wear metric and the mapped group with the lowest wear metric. If the lowest wear metric from the unmapped group is not greater than the lowest wear metric from the mapped group, the hybrid wear leveling management component 113 can swap the physical locations of the unmapped group with the lowest wear metric and the unmapped group with the highest wear metric.

[0044] Figure 4 is a flow chart of an example method 400 for performing hybrid wear leveling for in-place data replacement media according to some embodiments of the present disclosure. The method 400 may be performed by processing logic, which may include hardware (e.g., a processing device, a circuit, dedicated logic, programmable logic, microcode, hardware of a device, an integrated circuit, etc.), software (e.g., instructions running or executed on a processing device), or a combination thereof. In some embodiments, the method 400 is performed by Figure 1 The hybrid wear leveling management component 113 of the embodiment of the present invention is executed. Although shown in a specific order or sequence, the order of the processes can be modified unless otherwise specified. Therefore, it should be understood that the illustrated embodiments are only examples, and the illustrated processes can be performed in a different order, and some processes can be performed in parallel. In addition, one or more processes can be omitted in various embodiments. Therefore, not all processes are required in every embodiment. Other process flows are possible.

[0045] At block 410, the processing device periodically initiates a first wear leveling operation using a direct mapping function at a first frequency on the data management unit of the memory component 112. The direct mapping function may be part of the wear leveling operation and may include any mathematical function, such as an algebraic function, that is complete in itself and does require the use of any additional information in order to determine the target physical location in the memory component 112A. The first wear leveling operation may occur at a first level using direct algebraic mapping at a first frequency, such as every five minutes or every 1000 host writes, which may be configurable to a specific implementation.

[0046] At block 420, the processing device applies a first logical index associated with data from one of the data units in the first level of the hierarchy to a direct mapping function to determine a physical index corresponding to a physical location on the memory component 112A. In one embodiment, the direct mapping function includes at least one of a swap function, a circular shift function, a linear function utilizing a base pointer value and a free pointer value, etc. At block 430, the processing device copies the data from the data management unit to the physical location.

[0047] At block 440, the processing device periodically initiates a second wear leveling operation using the indirect mapping on the group of data management units of the memory component 112A at a second frequency, wherein the second wear leveling operation is performed less frequently than the first wear leveling operation. With the indirect mapping, information needs to be added in order to target a target physical location on the memory component 112A. The second wear leveling operation may occur at a second frequency (e.g., every hour or every 10,000 host writes) using an indirect fully associative mapping implemented by a lookup table at a second level, which may be configurable to a specific implementation.

[0048] At block 450, the processing device copies the data from the group of data management units to an available physical location on the memory component 112 A. This physical location may be determined in any number of ways, such as the location with the lowest write count, the location least recently accessed, and so on.

[0049] At block 460, the processing device records the mapping of the logical indexes associated with the data from the group of data management units to the available physical locations in a lookup table. After copying the data, the hybrid wear leveling management component 113 records the mapping of the logical indexes associated with the data from the group of data management units to the available physical locations in a lookup table (e.g., lookup table 240). In one embodiment, the lookup table is maintained in the local memory 119 of the controller 115. In other embodiments, the mapping may be maintained in some other type of data structure, such as an array, a linked list, etc.

[0050] Figure 5 An example machine of a computer system 500 is illustrated within which a set of instructions for causing the machine to perform any one or more of the methodologies discussed herein may be executed. In some embodiments, the computer system 500 may correspond to a computer system that includes, is coupled to, or utilizes a memory subsystem (e.g., Figure 1 A host system (eg, a memory subsystem 110) of Figure 1 host system 120) or can be used to perform operations of the controller (e.g., execute an operating system to perform operations corresponding to Figure 1In some embodiments, the machine may be connected (e.g., using a network) to other machines in a LAN, an intranet, an extranet, and / or the Internet. The machine may operate in the capacity of a server or a client machine in a client-server network environment, or in the capacity of a peer machine in a peer-to-peer (or distributed) network environment, or in the capacity of a server or a client machine in a cloud computing infrastructure or environment.

[0051] The machine may be a personal computer (PC), a tablet PC, a set-top box (STB), a personal digital assistant (PDA), a cellular phone, a network appliance, a server, a network router, a switch or a bridge, or any machine capable of executing (sequentially or otherwise) a set of instructions that specify actions to be taken by the machine. In addition, while a single machine is described, the term "machine" should also be construed to include any collection of machines that individually or collectively execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein.

[0052] The example computer system 500 includes a processing device 502, a main memory 504 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM), such as synchronous DRAM (SDRAM) or Rambus DRAM (RDRAM)), etc.), a static memory 506 (e.g., flash memory, static random access memory (SRAM), etc.), and a data storage system 518, which communicate with each other via a bus 530.

[0053] The processing device 502 represents one or more general-purpose processing devices, such as a microprocessor, a central processing unit, and the like. More specifically, the processing device may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, or a processor that implements other instruction sets, or a processor that implements a combination of instruction sets. The processing device 502 may also be one or more special-purpose processing devices, such as an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), a network processor, and the like. The processing device 502 is configured to execute instructions 526 for performing the operations and steps discussed herein. The computer system 500 may further include a network interface device 508 to communicate over a network 520.

[0054] The data storage system 518 may include a machine-readable storage medium 524 (also referred to as a computer-readable medium) on which is stored one or more sets of instructions 526 or software embodying any one or more of the methodologies or functions described herein. The instructions 526 may also reside, completely or at least partially, within the main memory 504 and / or within the processing device 502 during execution thereof by the computer system 500, the main memory 504 and the processing device 502 also constituting machine-readable storage media. The machine-readable storage medium 524, the data storage system 518, and / or the main memory 504 may correspond to Figure 1 Memory subsystem 110.

[0055] In one embodiment, instructions 526 include implementing instructions corresponding to a component (e.g., Figure 1 The hybrid wear leveling management component 113 of the present invention may be used to perform the functional instructions of the hybrid wear leveling management component 113 of the present invention. Although the machine-readable storage medium 524 is shown as a single medium in the example embodiment, the term "machine-readable storage medium" should be considered to include a single medium or multiple media storing one or more sets of instructions. The term "machine-readable storage medium" should also be considered to include any medium capable of storing or encoding a set of instructions for execution by a machine and causing the machine to perform any one or more of the methods of the present disclosure. The term "machine-readable storage medium" should therefore be considered to include, but not limited to, solid-state memory, optical media, and magnetic media.

[0056] Some portions of the previous detailed description have been presented in terms of algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations are the means by which those skilled in the art of data processing can most effectively convey the substance of their work to other persons skilled in the art. An algorithm is here and generally considered to be a self-consistent sequence of operations that result in a desired result. An operation is one that requires physical manipulation of physical quantities. These quantities are typically, but not necessarily, in the form of electrical or magnetic signals that can be stored, combined, compared, and otherwise manipulated. At times, primarily for general reasons, it has proven convenient to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.

[0057] It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. The present disclosure may refer to the actions and processes of a computer system or similar electronic computing device that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system's memories or registers or other such information storage systems.

[0058] The present disclosure also relates to an apparatus for performing the operations described herein. This apparatus may be specially constructed for the intended purpose, or it may comprise a general purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a computer-readable storage medium, such as, but not limited to, any type of disk, including floppy disks, optical disks, CD-ROMs and magnetic optical disks, read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, or any type of medium suitable for storing electronic instructions, which are respectively coupled to a computer system bus.

[0059] The algorithms and displays presented herein are not inherently related to any particular computer or other device. Various general purpose systems may be used with programs according to the teachings herein, or it may prove convenient to construct more specialized devices to perform the methods. The structures of a variety of these systems will be presented as set forth in the description below. In addition, the present disclosure is described without reference to any particular programming language. It will be appreciated that a variety of programming languages ​​may be used to implement the teachings of the present disclosure as described herein.

[0060] The present disclosure may be provided as a computer program product or software, which may include a machine-readable medium having stored thereon instructions that can be used to program a computer system (or other electronic device) to perform a process according to the present disclosure. A machine-readable medium includes any mechanism for storing information in a form readable by a machine (e.g., a computer). In some embodiments, a machine-readable (e.g., computer-readable) medium includes a machine (e.g., computer) readable storage medium, such as a read-only memory ("ROM"), a random access memory ("RAM"), a magnetic disk storage medium, an optical storage medium, a flash memory component, etc.

[0061] In the foregoing description, embodiments of the present disclosure have been described with reference to specific example embodiments thereof. It will be apparent that various modifications may be made to the present disclosure without departing from the broader spirit and scope of the embodiments of the present disclosure as set forth in the appended claims. Therefore, the description and drawings should be viewed in an illustrative sense rather than a restrictive sense.

Claims

1. A system comprising: Memory device; and A processing device operatively coupled to the memory device to perform operations comprising: periodically performing a first wear leveling operation using a direct mapping function at a first frequency on a data management unit of the memory device; and periodically performing a second wear leveling operation using an indirect mapping at a second frequency on a group of data management units of the memory device, wherein the second wear leveling operation is performed less frequently than the first wear leveling operation, and wherein periodically performing the second wear leveling operation comprises: identifying an available physical location among a plurality of physical locations on the memory device, the available physical location having at least one of a lowest write count or least recently accessed among the plurality of physical locations on the memory device; copying data from one of the groups of data management units to the available physical location; and An entry in a lookup table is updated, the entry corresponding to the logical index associated with the one of the groups of data management units and indicating a physical index corresponding to the available physical location.

2. The system of claim 1, wherein the memory device comprises a cross-point array of non-volatile memory cells.

3. The system of claim 1 , wherein periodically performing the first wear leveling operation using the direct mapping function comprises: applying a first logical index associated with data from the data management unit to the direct mapping function to determine a physical index corresponding to a physical location on the memory device; and The data from the data management unit is copied to the physical location. The system of claim 3 , wherein the direct mapping function comprises a swap function.

5. The system of claim 3, wherein the direct mapping function comprises a cyclic shift function.

6. The system of claim 3, wherein the direct mapping function comprises a linear function utilizing a base pointer value and a free pointer value.

7. The system of claim 1, wherein the first frequency and the second frequency are based on a time period that has passed since a previous wear leveling operation was performed.

8. The system of claim 1, wherein the first frequency and the second frequency are based on a number of data write operations performed by a host machine to the memory device since performing a previous wear leveling operation.

9. A method comprising: periodically performing a first wear leveling operation using a direct mapping function at a first frequency on a data management unit of the memory device; and periodically performing a second wear leveling operation using an indirect mapping at a second frequency on a group of data management units of the memory device, wherein the second wear leveling operation is performed less frequently than the first wear leveling operation, and wherein periodically performing the second wear leveling operation comprises: identifying an available physical location among a plurality of physical locations on the memory device, the available physical location having at least one of a lowest write count or least recently accessed among the plurality of physical locations on the memory device; copying data from one of the groups of data management units to the available physical location; and An entry in a lookup table is updated, the entry corresponding to the logical index associated with the one of the groups of data management units and indicating a physical index corresponding to the available physical location.

10. The method of claim 9, wherein the memory device comprises a cross-point array of non-volatile memory cells.

11. The method of claim 9, wherein periodically performing the first wear leveling operation using the direct mapping function comprises: applying a first logical index associated with data from the data management unit to the direct mapping function to determine a physical index corresponding to a physical location on the memory device; and The data from the data management unit is copied to the physical location.

12. The method of claim 11, wherein the direct mapping function comprises a swap function.

13. The method of claim 11, wherein the direct mapping function comprises a cyclic shift function.

14. The method of claim 11, wherein the direct mapping function comprises a linear function utilizing a base pointer value and an idle pointer value.

15. The method of claim 9, wherein the first frequency and the second frequency are based on a time period that has elapsed since a previous wear leveling operation was performed.

16. The method of claim 9, wherein the first frequency and the second frequency are based on a number of data write operations performed by a host machine on the memory device since performing a previous wear leveling operation.

17. A non-transitory computer-readable storage medium comprising instructions that, when executed by a processing device, cause the processing device to perform operations comprising: periodically performing a first wear leveling operation using a direct mapping function at a first frequency on a data management unit of the memory device; and periodically performing a second wear leveling operation using an indirect mapping at a second frequency on a group of data management units of the memory device, wherein the second wear leveling operation is performed less frequently than the first wear leveling operation, and wherein periodically performing the second wear leveling operation comprises: identifying an available physical location among a plurality of physical locations on the memory device, the available physical location having at least one of a lowest write count or least recently accessed among the plurality of physical locations on the memory device; copying data from one of the groups of data management units to the available physical location; and An entry in a lookup table is updated, the entry corresponding to the logical index associated with the one of the groups of data management units and indicating a physical index corresponding to the available physical location.

18. The non-transitory computer-readable storage medium of claim 17, wherein the memory device comprises a cross-point array of non-volatile memory cells.

19. The non-transitory computer-readable storage medium of claim 17, wherein periodically performing the first wear leveling operation using the direct mapping function comprises: applying a first logical index associated with data from the data management unit to the direct mapping function to determine a physical index corresponding to a physical location on the memory component; and The data from the data management unit is copied to the physical location.

20. The non-transitory computer-readable storage medium of claim 17, wherein the first frequency and the second frequency are based on at least one of a time period that has passed since a previous wear leveling operation was performed or a number of data write operations performed by a host machine on the memory device since the previous wear leveling operation was performed.