Storage device and data processing method
By combining flexible data placement and dynamic cache strategies in storage devices, the problem of low space utilization of cache media in the prior art is solved, and performance improvement in multi-VM environments is achieved.
Patent Information
- Application Number
- CN202410653024.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-01-19
- Filing Date
- 2024-05-24
- Publication Date
- 2024-12-03
AI Technical Summary
When existing storage devices deal with workloads with low cache hit rates, the space of cache media cannot be effectively utilized, resulting in latency and performance problems. Especially in multi-VM environments, cache policies are difficult to balance the cache policy for each virtual machine. Hit rate.
By introducing a flexible data placement scheme in the storage device, data placement information is used to store data in storage media and cache media at the same time, and cache policies and configuration information are dynamically adjusted according to different data access modes to improve the cache hit rate and utilization.
It realizes effective utilization of cache space under low hit rate workloads, reduces latency, improves performance, and adapts to different data access modes through flexible data placement and cache configuration strategies, and improves the overall performance of storage devices.
Smart Images

Figure CN119065591A_ABST
Abstract
Description
[0001] This application claims priority to and the benefit of U.S. Provisional Patent Application No. 63 / 469,823, filed on May 30, 2023, and U.S. Non-Provisional Patent Application No. 18 / 418,066, filed on January 19, 2024, which are incorporated by reference. Technical Field
[0002] The present disclosure relates generally to storage, and more particularly to storage devices and data processing methods. Background Art
[0003] The storage device may include a storage medium configured to store data received at the storage device. The storage device may also include a cache for storing copies of the data stored in the storage medium. An application or other user may access the storage device using one or more communication interfaces, communication protocols, etc.
[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the principles of the invention and therefore it may contain information that does not constitute prior art. Summary of the invention
[0005] An apparatus may include a first memory medium, a second memory medium, and at least one control circuit, the at least one control circuit being configured to receive placement information of data, store the data in a portion of the first memory medium based on the placement information, and store the data in a portion of the second memory medium based on the placement information. The at least one control circuit may be configured to receive a request for accessing the data from the portion of the first memory medium, and access the data from the portion of the second memory medium based on the request. The at least one control circuit may be configured to modify the portion of the first memory medium to an available state based on an allocation state. The portion of the first memory medium may be a first portion of the first memory medium, the portion of the second memory medium may be a first portion of the second memory medium, the placement information may be first placement information, the data may be first data, and the at least one control circuit may be configured to receive second placement information of the second data, store the second data in a second portion of the first memory medium based on the second placement information, and store the second data in a second portion of the second memory medium based on the second placement information. The at least one control circuit may be configured to receive configuration information, configure the first portion of the first memory medium based on the configuration information, and configure the first portion of the second memory medium based on the configuration information. The at least one control circuit may be configured to receive first configuration information and configure a first portion of the second memory medium based on the first configuration information. The at least one control circuit may be configured to receive second configuration information and configure a second portion of the second memory medium based on the second configuration information. The at least one control circuit may be configured to receive a request for performing an operation associated with the portion of the first memory medium and perform the operation using the portion of the second memory medium based on the request. The operation may include copying at least a portion of the data from the portion of the second memory medium to the portion of the first memory medium. The operation may include modifying a state of at least a portion of the portion of the second memory medium. The at least one control circuit may be configured to perform an access operation based on a result of a first operation for accessing the portion of the second memory medium, the access operation including performing a second operation for accessing the portion of the first memory medium and performing a third operation for accessing the portion of the second memory medium. The result of the first operation for accessing the portion of the second memory medium may include reading at least a portion of the data from the portion of the second memory medium. The at least one control circuit may also be configured to perform the access operation based on a result of a fourth operation for accessing the portion of the second memory medium.A result of the first operation for accessing the portion of the second storage medium may include reading at least a portion of the data from the portion of the second storage medium, and a result of the fourth operation for accessing the portion of the second storage medium may include reading at least a portion of the data from the portion of the first storage medium.
[0006] An apparatus may include a first memory medium, a second memory medium, and at least one control circuit, the at least one control circuit being configured to receive first configuration information for configuring a portion of the first memory medium based on placement information of data, receive second configuration information for configuring a portion of the second memory medium based on the placement information of data, receive a request for accessing the data from the portion of the first memory medium, and access the data from the portion of the second memory medium based on the request. The placement information of the data may be first placement information of the first data, the portion of the first memory medium may be a first portion of the first memory medium, the portion of the second memory medium may be a first portion of the second memory medium, and the at least one control circuit may be configured to receive third configuration information for configuring the second portion of the first memory medium based on the second placement information of the second data, and receive fourth configuration information for configuring the second portion of the second memory medium based on the second placement information of the second data. The at least one control circuit may be configured to modify the portion of the first memory medium to an available state based on an allocation state.
[0007] A method may include storing data in a portion of a first storage medium based on placement information, storing the data in a portion of a second storage medium based on the placement information, receiving a request to access the data from the portion of the first storage medium, and accessing the data from the portion of the second storage medium based on the request. The portion of the first storage medium may be a first portion of the first storage medium, the portion of the second storage medium may be a first portion of the second storage medium, the placement information may be first placement information, the data may be first data, the request may be a second request, and the method may further include storing second data in a second portion of the first storage medium based on second placement information, storing the second data in a second portion of the second storage medium based on the second placement information, receiving a second request to access the second data from the second portion of the first storage medium, and accessing the second data from the second portion of the second storage medium based on the second request. The first portion of the second storage medium may operate based on first configuration information, and the second portion of the second storage medium may operate based on second configuration information. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] The drawings are not necessarily drawn to scale, and throughout the drawings, for illustrative purposes, elements of similar structure or function may generally be represented by the same reference numerals or portions thereof. The drawings are intended only to facilitate the description of the various embodiments described herein. The drawings do not describe every aspect of the teachings disclosed herein and do not limit the scope of the claims. In order to prevent the drawings from being obscure, not all components, connections, etc. may be shown, and not all components may have reference numerals. However, the pattern of the component configuration can be easily seen from the drawings. The drawings, together with the specification, illustrate example embodiments of the present disclosure and, together with the specification, are used to explain the principles of the present disclosure.
[0009] Figure 1 An embodiment of a flexible data placement scheme for storage media according to disclosed example embodiments is shown.
[0010] Figure 2 An embodiment of a caching scheme for a storage medium according to a disclosed example embodiment is shown.
[0011] Figure 3 An example embodiment of a caching scheme for a storage medium according to the disclosed example embodiments is shown.
[0012] Figure 4 Another example embodiment of a caching scheme for a storage medium according to the disclosed example embodiments is shown.
[0013] Figure 5 An example embodiment of a cache medium according to a disclosed example embodiment is shown.
[0014] Figure 6 An embodiment of a cache flushing method based on a data placement scheme according to a disclosed example embodiment is shown.
[0015] Figure 7 An example embodiment of a cache access method based on a data placement scheme according to a disclosed example embodiment is shown.
[0016] Figure 8 An example embodiment of a host device according to a disclosed example embodiment is shown.
[0017] Fig. 9 An example embodiment of an apparatus according to a disclosed example embodiment is shown.
[0018] Fig.10 An embodiment of a method for operating a cache using a data placement scheme according to a disclosed example embodiment is shown. DETAILED DESCRIPTION
[0019] The storage device may include a device cache, which may reduce the latency of some access operations, for example, by storing copies of frequently accessed data in a cache medium, which may have a relatively low access latency compared to the storage medium. Therefore, based on receiving a request to access data in the storage medium, the storage device may serve the request by accessing the copy stored in the cache medium. However, some workloads may have a relatively low cache hit rate, and therefore may not be able to effectively use the cache medium space. For example, a workload that accesses data with small locality (e.g., a relatively small amount of data that may be randomly dispersed in different locations) in an application (such as database operations, artificial intelligence, machine learning, etc.) may have a relatively low cache hit rate. In addition, if multiple virtual machines (VMs) access the same storage device, the cache may be configured to use a cache strategy that may increase the hit rate of one VM but reduce the hit rate of other VMs. Therefore, some parts of the device cache may operate in a wasteful manner.
[0020] In a caching scheme according to an example embodiment disclosed, placement information for data may be used to store data in a portion of a storage medium and a corresponding portion of a cache medium. For example, a device may configure one or more portions (e.g., physical portions) of a storage medium as one or more placement units (a placement unit may also be referred to as a data placement unit). The device may also configure one or more portions of the cache medium to operate as one or more caches for one or more corresponding portions of the storage medium. For example, a device may store in a portion of the cache medium a copy of data that may be (e.g., previously, currently, and / or eventually, depending on a cache policy) stored in a corresponding portion of the storage medium.
[0021] In some embodiments, the device may receive data and placement information for the data (e.g., from a user). The device may store the data in a portion of a storage medium based on the placement information for the data. The device may also store the data (e.g., a copy of the data) in a portion of a cache medium, which may be configured as a cache for a corresponding portion of the storage medium. Thus, in some embodiments, the device may use the placement information for the data to store the data (and / or a copy of the data) in a portion of the storage medium and / or a corresponding portion of the cache medium.
[0022] In some embodiments, a device may receive a request to access data that may be (e.g., previously, currently, and / or eventually) stored in a portion of a storage medium that may be determined by placement information for the data. The device may access the requested data in the portion of the storage medium. Alternatively or additionally, the device may access the requested data in a corresponding portion of a cache medium.
[0023] Examples of data placement units may include recycling groups and / or recycling units that can be used in a flexible data placement (FDP) scheme. The FDP scheme may store data of a user (e.g., an application) in a portion of a physical storage medium, which may be implemented, for example, using one or more recycling groups (and / or recycling units) that can be erased as one or more units (e.g., one or more erase blocks). In some embodiments, the storage medium may be erased by modifying the storage medium to a state that makes the storage medium available for storing data. For example, a storage medium implemented using a single-level cell (SLC) and non-NAND (NAND) flash memory may be erased by setting the state of a cell in an erase block to a logical one ("1"). Data may be stored in an erased cell by programming a specific cell in an erase block to a logical zero ("0").
[0024] In an FDP scheme, one or more reclaim groups and / or reclaim units may be identified, for example, using one or more data placement instructions, placement identifiers, reclaim group identifiers, placement handles, reclaim unit handles, reclaim unit handle identifiers, etc. Thus, in embodiments where one or more data placement units may be implemented (e.g., in an FDP implementation) with one or more reclaim groups and / or reclaim units, placement information for data to be stored in the one or more data placement units may include one or more data placement instructions, placement identifiers, reclaim group identifiers, placement handles, reclaim unit handles, reclaim unit handle identifiers, etc.
[0025] In some embodiments, a second portion of the device cache may be configured to cache data for a second portion of the storage medium, and the second portion of the storage medium may be configured as a second placement unit (e.g., configured as a second recycling group). Thus, different cache policies (e.g., configuration parameters (such as cache size, placement policy, replacement policy, write policy, etc.)) may be used for different portions of the cache corresponding to different portions of the storage medium. Depending on the implementation details, this may improve one or more cache operating parameters (such as hit rate, capacity utilization, latency, etc.) for one or more of the portions of the cache corresponding to the portions of the storage medium.
[0026] Additionally or alternatively, some embodiments may implement one or more policies for cache flushing, bypassing, etc. based on the configuration of a portion of the device cache and a corresponding portion of the storage medium. For example, a device may invalidate data in one or more portions of the device cache corresponding to one or more portions of the storage medium, e.g., based on a request received from a user of the one or more portions of the device cache. Depending on the implementation details, such embodiments may free up cache space for use by other users, portions of the storage medium, etc., thereby improving cache utilization and / or other performance parameters.
[0027] Additionally or alternatively, some embodiments may implement one or more serial and / or parallel access mechanisms for one or more portions of a cache and / or corresponding storage space, which may be configured as a placement unit (e.g., configured as a reclaim group using an FDP scheme). For example, data with a relatively low probability of a cache hit may be accessed from a portion of the cache in parallel with a corresponding portion of the storage medium, while data with a relatively high probability of a cache hit may be accessed from a portion of the cache sequentially with a corresponding portion of the storage medium. Depending on the implementation details, such embodiments may reduce latency, increase bandwidth, and / or improve other performance parameters.
[0028] The disclosure includes many principles related to using a cache for a storage medium to access data. The principles disclosed herein may have independent utility and may be embodied individually, and not every embodiment may utilize every principle. In addition, the principles may also be embodied in various combinations, some of which may amplify some benefits of the various principles in a synergistic manner.
[0029] For purposes of illustration, some example embodiments may be described in the context of some specific implementation details, however, the principles are not limited to these or any other implementation details. For example, some embodiments may be described in the context of a storage device that may use the Compute Express Link (CXL) and / or Non-Volatile Memory Express (NVMe) protocols, however, other embodiments may be implemented with other types of devices (such as accelerators, memory expansion devices, memory buffers, etc.) using other types of protocols. As another example, some embodiments may be described in the context of a storage medium that may have portions based on an FDP scheme (e.g., based on recycling groups and / or recycling units), but other embodiments may utilize other types of data placement schemes in which data may be placed in one or more placement units, for example, based on the user of the data, the use of the data, the type of the data, etc. Therefore, one or more schemes according to the disclosed example embodiments may use data placement information for its intended purpose, for other purposes, and / or for achieving the same or similar purposes in different ways.
[0030] Figure 1 An embodiment of a flexible data placement scheme for storage media according to disclosed example embodiments is shown. Figure 1The flexible data placement (FDP) scheme shown in enables a user (such as a host, an operating system, an application, a process, a service, a virtual machine (VM), a VM manager, etc.) to arrange data into one or more physical recovery units (RUs) 160-0, 160-1, 160-2, 160-3 ... (which may be collectively and / or individually referred to as 160) and / or recovery groups (RGs) 162-0, 162-1, ..., 162-P-1 (which may be collectively and / or individually referred to as 162) in a storage device, where P may be a positive integer. The recovery unit 160 may be implemented with a portion of a physical storage medium 166 that may be erased as a unit (e.g., one or more erase blocks). The recovery group may include one or more recovery units. Depending on the implementation details, this may, for example, reduce write amplification by enabling a user to place data that may be deallocated at the same time in the same recovery group and / or recovery unit.
[0031] The flexible data placement scheme may use recovery unit handles (RUHs) 164-0, 164-1, 164-2, ..., 164-N-1 (which may be collectively and / or individually referred to as 164) to enable a user to specify one or more recovery units 160 in a storage device, and the storage device may write data to the one or more recovery units 160, where N may be a positive integer. For example, a user may send a write command to the storage device. The write command may specify the data to be written to the storage device. The write command may also include or provide a technique for indicating a recovery unit handle 164 to specify one or more recovery units 160, and the storage device may write data to the one or more recovery units 160. At the storage device, the recovery unit handle 164 may initially reference the first recovery unit 160, and the storage device may write the data specified by the user to the first recovery unit 160.
[0032] The user may send one or more additional write commands specifying additional data to be written to the storage device using the same reclaim unit handle 164. The storage device may begin writing the additional data to a first reclaim unit 160 referenced by the reclaim unit handle 164. When the first reclaim unit becomes full, the storage device may modify the reclaim unit handle 164 to reference a second reclaim unit 160, and the storage device may continue to write data to the second reclaim unit 160. Although the first reclaim unit 160 is no longer referenced by the reclaim unit handle 164 (which may be referred to as a dereferenced or previously referenced state), the first reclaim unit 160 may remain associated with the reclaim unit handle 164 to indicate that it is being written to using the reclaim unit handle 164.
[0033] Additionally or alternatively, Figure 1 The FDP scheme shown in can enable a user to send a write command that can indicate the recycling group 162 to which the write data can be written. Additionally or alternatively, Figure 1 The FDP scheme shown in FIG. 1 may use one or more parameters to implement the FDP scheme (such as FDP attributes (which may identify attributes of the FDP configuration), a default placement identifier list (PIDL), etc.).
[0034] Figure 2 An embodiment of a caching scheme for a storage medium according to a disclosed example embodiment is shown. Figure 2 The embodiments shown in can be implemented with a device 202, which can include a storage medium 204, a cache medium 206, and / or a control logic 208. The device 202 can communicate with a user via a communication connection 207, which can transmit requests 209 (e.g., read and / or write requests, load and / or store requests, etc.), data 211, placement information 213, configuration information 215, addresses, metadata, etc. to and / or from the device 202.
[0035] The storage medium 204 may include at least one portion that may be configured and / or operated as a data placement unit 212. The data placement unit 212 may be implemented with physical portions of the storage medium 204, and a data placement scheme may place data into the physical portions based on, for example, a user of the data (e.g., a host, an operating system, an application, a process, a service, a VM, a VM manager, etc.), usage of the data (e.g., data, metadata, configuration data, error checking data, a buffer, a queue, a log, a journal, etc.), a type of the data (e.g., integer, real, character, list, table, array, audio, video, etc.), etc. The data placement unit 212 may be implemented, for example, with physical portions of the storage medium 204 (e.g., pages, blocks, sectors, rows, erase blocks, tracks (e.g., shingled magnetic recording (SMR) tracks), reclaim groups, reclaim units, etc.).
[0036] The cache medium 206 may include at least one portion of a cache area 214 (which may also be referred to as a cache) that may be configured for the data placement unit 212. The cache area 214 may be configured with one or more cache parameters (such as cache size, cache level (e.g., in a tiered cache scheme), placement policy, replacement policy, write policy, etc.). Examples of usable cache placement policies may include direct mapping, N-way set associative (or N-way set associative) (e.g., 2-way set associative, 4-way set associative, 8-way set associative, etc.), fully associative, etc. Examples of usable cache replacement policies may include least recently used (LRU), least frequently used (LFU), random replacement, etc., as well as replacement policies based on artificial intelligence (AI), machine learning (ML), etc. Examples of usable write policies may include write-back, write-through, etc.
[0037] The data placement unit 212 may be configured and / or operated based on the configuration information, the placement information 213, and the like. For example, a user may provide the configuration information 215 to the control logic 208, which may use the configuration information 215 to configure (e.g., allocate) a portion of the storage medium 204 as the data placement unit 212. As another example, a user may provide the placement information 213 (e.g., with the request 209), which may request and / or instruct the control logic 208 to store the data 211 associated with the request 209 in the data placement unit 212. As yet another example, a user may provide the configuration information 215 to the control logic 208, which may use the configuration information 215 to delete (e.g., deallocate) the data placement unit 212 (e.g., if the user may no longer use the data placement unit 212).
[0038] The cache region 214 may be configured and / or operated based at least in part on the configuration information 215, the placement information 213, and the like received by the control logic 208. For example, a user may provide the configuration information 215 to the control logic 208, which may use the configuration information 215 to configure (e.g., allocate) a portion of the cache medium 206 as the cache region 214. As another example, a user may provide the configuration information 215 to the control logic 208, which may use the configuration information 215 to associate the cache region 214 with the data placement unit 212. Thus, based on receiving the request 209 to access the data placement unit 212, the control logic 208 may check the cache region 214 for a cache hit or miss, and access the cache region 214 and / or the data placement unit 212 accordingly. As yet another example, a user may provide configuration information 215 to control logic 208, which may use configuration information 215 to configure cache region 214 to implement one or more placement policies, replacement policies, write policies, and the like.
[0039] Additionally or alternatively, the cache region 214 may be configured and / or operated based, at least in part, on one or more predetermined, learned, default, internal, and / or similar configurations, rules, policies, and / or the like. For example, the control logic 208 may implement one or more default cache configurations that may allocate a cache region 214 to the data placement unit 212 (e.g., a cache region 214 having one or more predetermined placement, replacement, and / or write policies). In some embodiments, such a default cache configuration may be replaced based on configuration information 215 received for the data placement unit 212 (e.g., from a user). In some embodiments, such a default cache configuration may be replaced or modified based on learned information (e.g., by monitoring access to the cache region 214 and / or the data placement unit 212 to determine one or more access patterns). In some embodiments, one or more default cache configurations may be determined by a user.
[0040] Depends on implementation details, Figure 2 The cache scheme shown in can improve one or more cache operating parameters (such as hit rate, capacity utilization, latency, etc.), free up cache space for use by other users, portions of storage media, etc., thereby improving cache utilization and / or other performance parameters, and / or improving one or more other performance parameters.
[0041] Any of the storage media disclosed herein, including the storage medium 204, may be implemented with any type of non-volatile memory medium, which may be used as, for example, a storage medium based on solid-state media, magnetic media, optical media, etc. For example, in some embodiments, the storage medium may be implemented with NAND flash memory, persistent memory (such as a cross-grid non-volatile memory), memory with bulk resistance change, phase change memory (PCM), etc., or any combination thereof.
[0042] The cache media disclosed herein, including the cache media 206, may be implemented by any available volatile memory media, such as dynamic random access memory (DRAM), static random access memory (SRAM), etc. Additionally or alternatively, any of the cache media disclosed herein may be implemented by any type of non-volatile memory media based on solid-state media, magnetic media, optical media, etc., or any combination thereof, as described above.
[0043] In some embodiments, storage medium 204 and cache medium 206 may be implemented with media that may have one or more relative characteristics that may make one or both of them more suitable for their respective functions. For example, in some embodiments, storage medium 204 may be implemented with NAND flash memory that may have relatively high capacity, low cost, etc., while cache medium 206 may be implemented with DRAM that may have relatively low access latency. As another example, in some embodiments, storage medium 204 may be implemented with magnetic media that may have relatively high capacity, low cost, etc., while cache medium 206 may be implemented with NAND flash memory that may have relatively low access latency.
[0044] Any of the communication connections disclosed herein, including the communication connection 207, may be implemented with any interconnect and / or storage interface and / or protocol, such as PCIe, NVMe, NVMe key-value (NVMe-KV), direct memory access (DMA), serial ATA (SATA), small computer system interface (SCSI), serial attached SCSI (SAS), compute express link (CXL), and / or one or more CXL protocols (such as CXL.mem, CXL.cache and / or CXL.io), Gen-Z, consistent accelerator processor interface (CAPI), cache coherent interconnect for accelerators (CCIX), etc., or any combination thereof. Alternatively or additionally, any of the communication connections disclosed herein, including the communication connection 207, may be implemented with any network interface and / or protocol, such as Ethernet, transmission control protocol / internet protocol (TCP / IP), remote direct memory access (RDMA), RDMA over converged Ethernet (RoCE), Fibre Channel, InfiniBand (IB), iWARP, NVMe- over fabric (NVMe-oF), etc., or any combination thereof. Alternatively or additionally, any of the communication connections disclosed herein including communication connection 207 may be implemented with any control and / or management interface including an NVMe management interface (NVMe-MI), a system management bus (SMBus), or the like.
[0045] Any of the devices disclosed herein, including device 202, may be implemented in any form (such as a storage device, an accelerator, a graphics processor (GPU), a network interface card, and / or a network interface controller (NIC), etc.) using any connector configuration (such as PCIe, SATA, SCSI, SAS, M.2, U.2, U.3, etc.), any physical form factor including one or more form factors for storage devices (e.g., solid state drives (SSDs), hard disk drives (HDDs), etc.) and / or other devices (such as peripheral component interconnect express (PCIe) add-in cards, 3.5-inch drives, 2.5-inch drives, 1.8-inch drives, M.2 drives, U.2 and / or U.3 drives, enterprise and data center SSD form factor (EDSFF) drives, any of the SFF-TA-100X form factors (e.g., SFF-TA-1002), NF1, etc.). In some embodiments, the device may be implemented in any other form (e.g., as a collection of one or more components on a circuit board (e.g., a circuit board integrated into a server motherboard, backplane, midplane, etc.)).
[0046] Any of the devices disclosed herein, including device 202, may be used in conjunction with one or more personal computers, smart phones, tablet computers, servers, server chassis, server racks, data rooms, data centers, edge data centers, mobile edge data centers, and / or any combination thereof.
[0047] Figure 3 An example embodiment of a caching scheme for a storage medium according to the disclosed example embodiments is shown. Figure 3 The embodiment shown in FIG. 1 can be used, for example, to implement Figure 2 In the embodiments shown in , similar elements may be indicated by reference numerals ending with and / or containing the same numbers, letters, etc.
[0048] Reference Figure 3 , the device 302 may include a storage medium 304, a cache medium 306, and / or a control logic 308. The device 302 may communicate via a communication connection using a communication interface 309, which may implement any interconnect, storage, network, management, etc., interface, protocol, etc. described above. Requests and / or data for accessing the storage medium 304 may be sent and / or received, for example, via a command and / or data (command / data) path 311.
[0049] The control logic 308 may include storage and / or cache access logic 316 (which may also be referred to as access logic) that may implement a cache scheme for the storage medium 304. For example, the access logic 316 may configure and / or operate a first portion and a second portion of the storage medium 304 as a data placement unit 362-0 and a data placement unit 362-1, which in this example may be implemented with reclaim group RG0 and reclaim group RG1, respectively. The access logic 316 may configure and / or operate reclaim group RG0 and / or reclaim group RG1 based on FDP configuration information 318, which may be received from a user, for example, via the communication interface 309.
[0050] The access logic 316 may also configure and / or operate the first and second portions of the cache medium 306 as cache areas 314-0 and cache areas 314-1 (which may also be referred to as caches, e.g., Cache 0 and Cache 1) for the reclaim group RG0 and the reclaim group RG1, respectively. The access logic 316 may configure and / or operate Cache 0 and / or Cache 1 based on cache configuration information 320, which may be received from a user, e.g., via the communication interface 309. For example, in some embodiments, the cache configuration information 320 may specify a first configuration for Cache 0, the first configuration for Cache 0 including a cache size, a placement strategy, a replacement strategy, a write strategy, etc., and the access logic 316 may use the first configuration for Cache 0 when processing a request to access the first reclaim group RG0. In such an embodiment, the cache configuration information 320 may also specify a second configuration for Cache 1, where the second configuration for Cache 1 includes cache size, placement strategy, replacement strategy, write strategy, etc. The access logic 316 may use the second configuration for Cache 1 when processing a request to access the second recycling group RG1.
[0051] Additionally or alternatively, the access logic 316 may configure and / or operate one or more of the caches 314-0 and / or caches 314-1 (which may also be referred to as cache regions) based, at least in part, on one or more predetermined, learned, default, internal, and / or similar configurations, rules, policies, etc. Additionally or alternatively, the access logic 316 may configure and / or operate one or more of the caches 314-0 and / or caches 314-1 based, at least in part, on at least some of the FDP configuration information 318. For example, in some embodiments, the access logic 316 may configure and / or operate one or more of the caches 314-0 and / or caches 314-1 based on a default configuration (e.g., in the absence of cache configuration information 320 for one or more of the caches 314-0 and / or caches 314-1).
[0052] Configuring and / or using multiple caches 314-0 and 314-1 for corresponding data placement units 362-0 and data placement units 362-1 can enable different cache configurations to be used for different data placement units. Depending on the implementation details, this can improve one or more cache operation parameters (such as hit rate, capacity utilization, latency, etc.) of one or more of caches 314-0 and / or caches 314-1.
[0053] As described above, one or more of reclaim group RG0 and / or reclaim group RG1 may be implemented with one or more reclaim units. In some embodiments, a reclaim unit may be implemented (at least from a user's perspective) as a logical representation of an underlying portion of a physical storage medium.
[0054] In some embodiments where the storage medium may be implemented at least in part with flash memory, a reclamation unit may refer to one or more erase blocks, non-volatile memory (NVM) devices (e.g., NVM dies), the like, or any combination thereof, and a reclamation group may refer to one or more reclamation units, one or more NVM device partitions (e.g., planes), one or more NVM devices (e.g., NVM dies), one or more storage devices (e.g., storage drives), the like, or any combination thereof.
[0055] In some embodiments where the storage media may be implemented at least in part with magnetic media (e.g., shingled magnetic recording (SMR) media), a recycling unit may refer to one or more shingled segments, zones, sectors, tracks, etc., or any combination thereof, and a recycling group may refer to one or more disks (e.g., drives), platters, tracks, zones, sectors, shingled segments, etc., or any combination thereof.
[0056] In some embodiments where the storage medium may be at least partially implemented with a storage class memory (e.g., magnetoresistive random access memory (MRAM), resistive random access memory (ReRAM), phase change memory (PCM), cross-grid non-volatile memory, memory with body resistance change, etc.), a recycling unit may refer to one or more storage bodies, programming groups, etc., or any combination thereof, and a recycling group may refer to one or more dies, storage bodies, programming groups, etc., or any combination thereof.
[0057] Figure 4 Another example embodiment of a caching scheme for a storage medium according to the disclosed example embodiments is shown. Figure 4 The embodiment shown in FIG. 1 can be used, for example, to implement Figure 2 and / or Figure 3 Any of the embodiments shown in , wherein similar elements may be indicated by reference numerals ending with and / or containing the same numbers, letters, etc.
[0058] Reference Figure 4, the storage device 402 may include a storage medium 404, a cache medium 406, a communication interface 409 and / or control logic, and the control logic may include one or more of the following elements: FDP configuration information 418, cache configuration information 420, an FDP manager 422, a flash translation layer (FTL) 424, a NAND interface layer (NIL) 426, a cache operation logic 428, a cache manager 430, a cache predictor 432, a first (e.g., NVMe) command manager 434 and / or a second (e.g., CXL) command manager 436.
[0059] The storage medium 404 may be implemented with NAND flash memory in this example, and the NAND flash memory may be arranged in channels CH0, CH1, ... CHN (N may be a positive integer (e.g., 7)) that may be controlled by control logic (e.g., one or more NAND channel controllers) in the NAND interface layer 426. The flash translation layer 424 may implement one or more mapping schemes for mapping a logical block address (PBA) to a physical block address (LBA) in the storage medium 404.
[0060] The FDP manager 422 may configure and / or operate one or more portions of the storage medium 404 as data placement units, which in this example may be implemented using reclaim groups based on the FDP configuration information 418. In this example, the FDP configuration information 418 may specify reclaim groups RG0, RG1, ..., RG5, each of which may include, for example, Figure 4 A specified number of recovery units 460 are shown in FIG.
[0061] The cache medium 406 in this example may be implemented using DRAM. In some embodiments, the cache manager 430 may use the cache configuration information 420 to configure and / or operate one or more portions of the cache medium 406 as caches 414-0, 414-1, ... (which may be collectively and / or individually referred to as 414, and which may also be referred to as cache regions) for one or more corresponding recycle groups RG0, RG1, ..., RG5, respectively. In this example, the cache configuration information 420 may specify caches Cache 0, Cache 1, ..., Cache 5 corresponding to the recycle groups RG0, RG1, ..., RG5, respectively.
[0062] Additionally or alternatively, cache manager 430 may configure and / or operate one or more caches 414 based, at least in part, on one or more predetermined, learned, default, internal, and / or similar configurations, rules, policies, etc. Additionally or alternatively, cache manager 430 may configure and / or operate one or more of caches 414 based, at least in part, on at least some of FDP configuration information 418. For example, in some embodiments, cache manager 430 may configure and / or operate one or more of caches 414 based on a default configuration (e.g., in the absence of cache configuration information 420 for one or more of caches 414).
[0063] Figure 4 The scheme shown in 404 may also include a host 401, and one or more applications 438-1, 438-2, ..., 438-M (which may be collectively and / or individually referred to as 438) may be running on the host 401. The host 401 may include a user-side FDP configuration logic 440, which may be used by a user (e.g., application 438 and / or one or more of an operating system, process, service, VM, VM manager, etc.) to generate at least a portion of FDP configuration information 418 and / or send at least a portion of FDP configuration information 418 to the storage device 402. The user-side FDP configuration logic 440 may be at least partially located in (e.g., distributed in) one or more of the application 438 and / or the operating system, process, service, VM, VM manager, etc.
[0064] The host 401 may include user-side cache configuration logic 442, which may be used by a user (e.g., application 438 and / or one or more of an operating system, process, service, VM, VM manager, etc.) to generate at least a portion of cache configuration information 420 and / or send at least a portion of cache configuration information 420 to the storage device 402. The user-side cache configuration logic 442 may be at least partially located in (e.g., distributed in) one or more of the application 438 and / or the operating system, process, service, VM, VM manager, etc. In some embodiments, the user-side cache configuration logic 442 may enable a user to adjust cache size, location, etc., for example, based on the size, location, etc. of an associated data placement unit.
[0065] In some embodiments, FDP configuration information 418 and / or cache configuration information 420 may be sent to storage device 402 as indicated by arrows 419 and / or 421, respectively, using one or more interfaces, protocols, etc. that may support configuration and / or management operations (such as, NVMe-MI, CXL.io, etc.).
[0066] The cache configuration information 420 may be stored, for example, using a data structure (such as a table, a linked list, a key-value store, etc.) having a structure such as Figure 4 , for identifying one or more recycling groups RG and corresponding cache configuration parameters. For example, Figure 4 The cache configuration information 420 shown in may include identifiers, indexes, etc. for one or more (e.g., each) of the reclaim groups RG0, RG1, ..., and corresponding parameters (such as placement strategy (e.g., the number of ways and / or sets for an N-way set associative cache), replacement strategy, write strategy, cache size, one or more base addresses of the cache 414-0 in the cache medium 406, etc.). Therefore, the cache configuration information 420 may include different cache configuration parameters for different reclaim groups, and thus different cache strategies for different reclaim groups may be implemented.
[0067] The cache predictor 432 may provide one or more indications, estimates, predictions, etc. to enable the cache manager 430 to implement one or more cache configurations, operations, policies, strategies, etc. based on one or more commands, requests, etc. received from the cache operation logic 428, the cache configuration information 420, observations of accesses to the storage medium 404 and / or the cache medium 406, etc. For example, the cache predictor 432 may determine the likelihood of a cache hit, a cache eviction, etc. to enable the cache manager 430 to implement a replacement strategy for the one or more caches 414. As another example, the cache predictor 432 may determine the likelihood of a cache hit to enable the cache manager 430 to implement serial and / or parallel data access operations for the one or more caches 414.
[0068] The cache operation logic 428 may implement one or more cache operations (such as a flush operation, a bypass operation, an access operation, etc.) for the one or more caches 414. One or more of the cache operations may be based on, for example, one or more commands that may be received by the cache operation logic 428 as indicated by arrow 429 using one or more interfaces, protocols, etc. (such as NVMe-MI, CXL.io, etc.) that may support configuration and / or management operations.
[0069] The first command manager 434 may process commands for accessing the storage medium 404, data to be stored in the storage medium 404, data retrieved from the storage medium 404, etc. In this example, the first command manager 434 may implement the NVMe protocol, which may enable a user (e.g., an application 438) to access the storage medium 404 as storage using a read and / or write (read / write) command that may specify a read and / or write address having one or more LBAs. One or more connections 444 utilizing the NVMe protocol may be used to send and / or receive read / write commands, write data, read data, etc. In some embodiments, the NVMe protocol may be implemented using an underlying CXL.io transport scheme.
[0070] In some embodiments, an access (e.g., read / write) request received by the first command manager 434 (e.g., from an application 438) may include placement information (e.g., a reclaim unit handle (RUH) that may identify a reclaim group and / or other placement information) to enable the FDP manager 422 to service the access request using a specific data placement unit (e.g., a reclaim unit, a reclaim group, etc.) specified in the FDP configuration information 418.
[0071] The FDP manager 422 may route the access request received by the first command manager 434 to the FTL 424, which may check the cache manager 430 to determine whether the cache 414 (e.g., Cache1) is associated with the reclaim group (e.g., RG1) associated with the access request, and if so, check for a cache hit. The cache manager 430 and / or the FDP manager 422 may process cache hits, cache misses, etc. based on one or more cache parameters associated with the reclaim group (e.g., RG1) in the cache configuration information 420. For example, in the case of a cache miss for a read command, the cache manager 430 and / or the FDP manager 422 may read the requested data from the corresponding reclaim group (e.g., RG1) in the storage medium 404, evict one or more cache lines from the corresponding cache 414 (e.g., Cache 1) based on the replacement policy specified for RG1 in the cache configuration information 420, and send the requested read data to the first command manager 434, which may send the requested read data to the user (e.g., application 438) that sent the read request. In some embodiments, the cache predictor 432 may update one or more data structures with information about the read access operation to learn one or more access patterns for, for example, subsequent predictions.
[0072] Some embodiments may include a second command manager 436 that may process commands for accessing the storage medium 404, data to be stored in the storage medium 404, data retrieved from the storage medium 404, etc. In this example, the second command manager 436 may implement a CXL protocol that may enable a user (e.g., an application 438) to access the storage medium 404 as a memory using load and / or store (load / store) commands that may specify a requested data location as a memory address (e.g., a memory address that may be mapped to a host management device memory (HDM) region in system memory). Load / store commands, load data, store data, etc. may be sent and / or received using one or more connections 446, for example, using the CXL.mem protocol.
[0073] therefore, Figure 4 The embodiment shown in can implement a dual mode (eg, dual access method) scheme, in which, as shown in FIG. Figure 4 As shown in , a user (e.g., application 438) can use CXL load / store commands to access data placement units (e.g., reclaim groups) as memory and / or use NVMe read / write commands to access data placement units (e.g., reclaim groups) as storage devices.
[0074] In some embodiments, the second command manager 436 may implement a memory-to-storage address translator (e.g., a CXL-to-LBA translator or a CXL2LBA translator) to convert memory addresses included in load / store commands into LBAs that the FDP manager 422 and / or FTL 424 may use to access requested data in the storage medium 404 in a manner similar to that described above with respect to access requests received by the first command manager 434.
[0075] In some embodiments, any or all of the FDP configuration information 418, cache configuration information 420, FDP manager 422, FTL 424, NAND interface layer 426, cache operation logic 428, cache manager 430, cache predictor 432, first command manager 434, and / or second command manager 436 may implement control logic and / or may be collectively and / or individually referred to as control logic. In some embodiments, any or all of the FDP manager 422, FTL 424, NAND interface layer 426, cache operation logic 428, cache manager 430, cache predictor 432, first command manager 434, and / or second command manager 436 may implement and / or may be collectively and / or individually referred to as storage and / or cache access logic (e.g., similar to Figure 3). In some embodiments, Figure 4 The storage device 402 shown in the figure may implement a configurable (and / or reconfigurable) cache architecture based on a data placement scheme (e.g., an FDP scheme). In some embodiments, any or all of the FDP configuration information 418, the cache configuration information 420, the FDP manager 422, the cache operation logic 428, the cache manager 430, the cache predictor 432, the first command manager 434, and / or the second command manager 436 may implement and / or may be collectively and / or individually referred to as a host interface layer (HIL).
[0076] Figure 5 An example embodiment of a cache medium according to a disclosed example embodiment is shown. Figure 5 The cache medium 506 shown in FIG. 5 may be used, for example, to implement Figure 4 Cache medium 406 is shown in , where similar elements may be indicated by reference numerals ending with and / or containing the same numbers, letters, etc.
[0077] Reference Figure 5 , the cache medium 506 may include one or more portions, the one or more portions being based on, for example, cache configuration information (such as, Figure 4 Cache configuration information 420 shown in FIG. 4 is configured for one or more corresponding data placement units (eg, Figure 4 , RG0, RG1, ..., RG5) are shown in FIG. 5 . Although the cache medium according to the disclosed example embodiments is not limited to any particular configuration, for the purpose of illustrating the disclosed principles, the cache medium is shown in FIG. Figure 5In the example shown in , Cache 0 may be configured as a cache with a direct mapping placement strategy and / or a random replacement strategy for RG0; Cache 1 may be configured as a cache with a direct mapping placement strategy and / or a random replacement strategy for RG1; Cache 2 may be configured as a cache with a 2-way set associative placement strategy and / or a least recently used (LRU) replacement strategy for RG2; Cache 3 may be configured as a cache with a 2-way set associative placement strategy and / or a least recently used replacement strategy for RG3; Cache 4 may be configured as a cache with a 4-way set associative placement strategy and / or a least frequently used (LFU) replacement strategy for RG4; and / or Cache 5 may be configured as a cache with a direct mapping placement strategy and / or a random replacement strategy for RG5. In one example, a cache may store data and tags corresponding to the data according to a placement strategy and / or a replacement strategy of the cache.
[0078] Figure 6 An embodiment of a cache flushing method based on a data placement scheme according to a disclosed example embodiment is shown. Figure 6 The method shown in can be used, for example, Figure 4 The cache operation logic 428, cache manager 430, cache configuration information 420, etc. shown in are implemented, where similar elements can be indicated by reference numerals ending with and / or containing the same numbers, letters, etc.
[0079] Reference Figure 6 , the method may begin at operation 648, where a cache flush request for a target data placement unit (e.g., in the form of a command received from a user by the cache operation logic 428) may be received. The cache flush request may include one or more LBAs, LBA ranges, one or more RUHs, and / or other information, the target data placement unit (e.g., a reclaim group) may be determined from the one or more LBAs, LBA ranges, one or more RUHs, and / or other information, and the cache flush method may be performed for the target data placement unit (e.g., a reclaim group).
[0080] At operation 650, the method may determine an address range (e.g., a page address range) of a target cache (e.g., Cache 0) (which may also be referred to as a cache region) corresponding to the target data placement unit (e.g., RG0) based on the target data placement unit (e.g., RG0) and / or using cache configuration information 620. At operation 652, the method may flush (e.g., copy) some or all of the data in the target cache region to the corresponding target data placement unit. In some embodiments, at operation 652, the method may also invalidate (e.g., mark as invalid) some or all of the data in the target cache region. In some embodiments, invalidating the data in the target cache region may free up (e.g., deallocate) memory and / or storage media used to implement the cache medium for use by other users. Depending on the implementation details, this may improve cache utilization and / or other performance parameters.
[0081] Figure 6 The method shown in may be used, for example, by a user who may realize that data that may be stored in a data placement unit (e.g., a reclaim group) may have little or no locality and, therefore, using the associated cache region may provide little, if any, benefit. Thus, in some embodiments, Figure 6 The method shown in can provide relatively fast cache space release.
[0082] In some embodiments, Figure 6 The method shown in can be used as part of a cache bypass method. For example, if an application (or other user) determines that the data it will access has little or no locality, the application can request (e.g., by sending a bypass command to the cache operation logic 428) a temporary or permanent bypass configuration in which one or more accesses of a reclaim group can bypass the corresponding cache region and directly access the reclaim group in the storage medium (e.g., without checking for a cache hit). In such a bypass configuration, the cache operation logic 428 can use Figure 6 The method shown in can flush some or all of the data in the target cache area to the corresponding target data placement unit and / or invalidate some or all of the data in the target cache area. Depending on the implementation details, this can release the memory and / or storage medium used to implement the cache medium for use by other users at least while using the bypass configuration.
[0083] Figure 7 An example embodiment of a cache access method based on a data placement scheme according to a disclosed example embodiment is shown. Figure 7 The method shown in can be used, for example, Figure 4The cache operation logic 428, cache manager 430, cache configuration information 420, etc. shown in are implemented, where similar elements can be indicated by reference numerals ending with and / or containing the same numbers, letters, etc.
[0084] In an example embodiment of a cache access method based on a data placement scheme, a request for accessing data from a portion of a first memory medium (e.g., a storage device) (e.g., a data placement unit corresponding to the data) is received, and an access operation for the request is performed based on cache hit information corresponding to a portion of a second memory medium (e.g., a cache) (e.g., a cache area corresponding to the data). The cache hit information is determined based on placement information of the data and / or cache configuration information corresponding to the portion of the second memory medium. In one example, in response to the cache hit information including a possible cache hit rate greater than or equal to a threshold, the data is accessed from the portion of the second memory medium, and in response to a cache miss occurring when accessing the data from the portion of the second memory medium, the data is accessed from the portion of the first memory medium. In another example, in response to the cache hit information including a possible cache hit rate less than a threshold, the data is accessed from the portion of the first memory medium and the data is accessed from the portion of the second memory medium in parallel.
[0085] Reference Figure 7 The method may begin at operation 754 where a read request (eg, via Figure 4 At operation 756, the method may determine cache hit information (such as a possible cache hit rate (e.g., estimated and / or predicted hit rate) of a target cache area corresponding to the target data placement unit) based on the target data placement unit and / or using the cache configuration information 720. The cache hit information may be, for example, Figure 4 The cache predictor 432 shown in FIG.
[0086] At operation 758, the method may determine an access technique to use for the read request. For example, if the cache hit information includes a relatively high possible cache hit rate, the method may proceed to operation 760, where the corresponding target cache area may be checked for a cache hit. In the event of a cache hit, the requested read data may be obtained from the corresponding target cache area and returned in response to the read request. However, if a cache miss is encountered, the method may proceed to operation 762, where the cache may be retrieved (e.g., by Figure 4The FDP manager 422 shown in FIG. 4 obtains the requested read data from the corresponding target data placement unit in the storage medium and returns the requested read data in response to the read request.
[0087] However, if at operation 758, the cache hit information includes a relatively low possible cache hit rate, the method may proceed to operation 764, where the corresponding target cache region may be (e.g., by Figure 4 ) is checked for cache hits, and the method may proceed to operation 766 where the cache manager 430 may be (e.g., via Figure 4 ) obtains the requested read data from the corresponding target data placement unit in the storage medium. In some embodiments, the target cache area may be checked for cache hits, and the target data placement unit may be accessed in parallel. (For example, operation 764 and operation 766 may overlap.) This may reduce the delay associated with obtaining the requested read data from the target data placement unit, for example, by reducing or eliminating the delay associated with waiting for the result of checking the target cache area for a hit. In the event of a relatively unlikely cache hit, the requested read data may be obtained from the target cache area and / or the target data placement unit. The requested read data may be obtained from whichever source (for example, the target cache area or the target data placement unit) that first provides the data, and the requested read data is returned in response to the read request. Depending on the implementation details, this may reduce access latency, increase cache medium and / or storage medium bandwidth and / or efficiency, etc.
[0088] Figure 7 The method shown in can be applied to read requests received in any manner (e.g., using Figure 4 , and / or using a second (e.g., CXL) command manager 436 (e.g., using a CXL.mem data path) to access the data placement unit as a storage device.
[0089] Figure 8 Example embodiments of a host device that may be used to implement any of the host and / or other user functionality disclosed herein are shown in accordance with the disclosed example embodiments. Figure 8 The host device 880 shown in FIG. 8 may include a processor 882 (which may include a memory controller 884 ), a system memory 886 , host logic 888 , and / or a communication interface 883 . Figure 8 Any or all of the components shown in can communicate via one or more system buses 885. In some embodiments, Figure 8One or more of the components shown in may be implemented using other components. For example, in some embodiments, host logic 888 may be implemented by processor 882 executing instructions stored in system memory 886 or other memory.
[0090] Host logic 888 may be used to implement host and / or other user functionality disclosed herein (e.g., Figure 4 Any of the described host 401, application 438, user-side FDP configuration logic 440 and / or user-side cache configuration logic 442, etc.) one or more.
[0091] Fig. 9 An example embodiment of a device that can be used to implement any of the device functions disclosed herein according to the disclosed example embodiments is shown. The device 990 may include a device controller 992, a media conversion layer 994 (e.g., FTL), a storage medium 996, a cache medium 998, a control logic 997, and a communication interface 993. Fig. 9 The components shown in FIG. 9 may communicate via one or more device buses 995. In some embodiments where flash memory may be used for some or all of storage media 996, media conversion layer 994 may be partially or completely implemented as a flash translation layer (FTL).
[0092] Control logic 997 may be used to implement the device functions disclosed herein (e.g., Figure 4 432, any or all of the FDP configuration information 418, cache configuration information 420, FDP manager 422, FTL 424, NAND interface layer 426, cache operation logic 428, cache manager 430, cache predictor 432, first command manager 434 and / or second command manager 436 described herein.
[0093] Fig.10 An embodiment of a method for operating a cache using a data placement scheme according to an example embodiment of the disclosure is shown. The method may begin at operation 1002. At operation 1004, the method may store data in a portion of a first memory medium based on placement information. For example, referring to Figure 2 , the control logic 208 may store the data 211 in the data placement unit 212 in the storage medium 204 based on the placement information 213. In one example, a portion of the first storage medium may be modified to an available state based on the allocation state. For example, in response to a portion of the first storage medium being allocated, a portion of the first storage medium may be modified to an available state.
[0094] Reference Fig.10At operation 1006, the method may store the data in a portion of the second storage medium based on the placement information. Figure 2 , the control logic 208 may store the data 211 in a cache area 214 in the cache medium 206 based on the placement information 213 .
[0095] Reference Fig.10 At operation 1008, the method may receive a request to access data from a portion of the first storage medium. Figure 2 , control logic 208 may receive a request 209 to access data 211 from a data placement unit 212 in storage medium 204 .
[0096] Reference Fig.10 At operation 1010, the method may access data from a portion of the second storage medium based on the request. Figure 2 , control logic 208 may access data 211 from cache area 214 in cache medium 206 . The method may end at operation 1012 .
[0097] Fig.10 The embodiments shown in and all other embodiments described herein are example operations and / or components. In some embodiments, some operations and / or components may be omitted and / or other operations and / or components may be included. In addition, in some embodiments, the time and / or spatial order of operations and / or components may be changed. Although some components and / or operations may be shown as separate components, in some embodiments, some components and / or operations shown separately may be integrated into a single component and / or operation, and / or some components and / or operations shown as a single component and / or operation may be implemented with multiple components and / or operations.
[0098] Any of the functions described herein, including any functions that can be implemented by a host, a device (e.g., a storage device), etc., or a combination thereof, can be implemented by hardware (e.g., circuitry), software, firmware, or any combination thereof that executes instructions stored in any type of memory, or any combination thereof, including, for example, the functions described above with respect to Figure 3 and / or Figure 4Any of the described storage and / or cache access logic 316, FDP configuration information 418, cache configuration information 420, FDP manager 422, FTL 424, NAND interface layer 426, cache operation logic 428, cache manager 430, cache predictor 432, first command manager 434 and / or second command manager 436, user-side FDP configuration logic 440 and / or user-side cache configuration logic 442, etc., hardware (e.g., circuitry), software, firmware, or any combination thereof including, for example, hardware and / or software combinational logic, sequential logic, timers, counters, registers, state machines, volatile memories (such as , DRAM and / or SRAM), non-volatile memory including flash memory, persistent memory (such as cross-grid non-volatile memory), memory with body resistance change, PCM, etc., one or more complex programmable logic devices (CPLD), field programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), processors (including one or more complex instruction set computer (CISC) processors (such as x86 processors) and / or reduced instruction set computer (RISC) processors (such as ARM processors)), graphics processing units (GPUs), neural processing units (NPUs), tensor processing units (TPUs)), data processing units (DPUs), etc. In some embodiments, one or more components may be implemented as a system on a chip (SOC).
[0099] Although some embodiments disclosed above have been described in the context of various implementation details, the disclosed principles are not limited to these or any other specific details. For example, although some functions have been described as being implemented by specific components, in other embodiments, functions may be distributed in different locations and between different systems and components with various user interfaces. Specific embodiments have been described as having specific processes, operations, etc., but these terms also cover embodiments in which specific processes, operations, etc. can be implemented with multiple processes, operations, etc. or multiple processes, operations, etc. can be integrated into a single process, step, etc. Reference to a component or element may refer only to a part of a component or element. For example, a reference to a block may refer to an entire block or one or more sub-blocks. Unless otherwise clear from the context, terms such as "first" and "second" are used in disclosures and claims and may only be used to distinguish the purpose of the elements they modify, and may not indicate any spatial or temporal order. In some embodiments, reference to an element may refer to at least a portion of an element, for example, "based on" may refer to "based at least in part on", etc. Reference to a first element may not imply the presence of a second element. The principles disclosed herein have independent practicality and may be implemented separately, and not every embodiment may utilize each principle. However, these principles can also be embodied in various combinations, some of which can amplify the benefits of the individual principles in a synergistic manner. According to the inventive principles disclosed in this patent, the various details and embodiments described above can be combined to produce additional embodiments.
[0100] Since the inventive principles of this patent disclosure may be modified in arrangement and detail without departing from the inventive concept thereof, such changes and modifications are considered to fall within the scope of the appended claims.
Claims
1. A storage device, include: a first storage medium; a second storage medium; as well as At least one control circuit configured to: Receive placement information for data; storing the data in a portion of a first storage medium based on the placement information; and The data is stored in a portion of a second storage medium based on the placement information.
2. The storage device according to claim 1, in, The at least one control circuit is configured to: receiving a request to access the data from the portion of the first storage medium; and The data is accessed from the portion of the second storage medium based on the request.
3. The storage device according to claim 1, in, The at least one control circuit is configured to modify the portion of the first memory medium to an available state based on the allocation state.
4. The storage device according to claim 1, in, The portion of the first memory medium is a first portion of the first memory medium, the portion of the second memory medium is a first portion of the second memory medium, the placement information is first placement information, the data is first data, and the at least one control circuit is configured to: receiving second placement information of second data; storing second data in a second portion of the first storage medium based on the second placement information; and Second data is stored in a second portion of the second storage medium based on the second placement information.
5. The storage device according to claim 1, in, The at least one control circuit is configured to: receiving first configuration information and second configuration information; configuring the portion of the first storage medium based on the first configuration information; and The portion of the second memory medium is configured based on the second configuration information.
6. The storage device according to claim 4, in, The at least one control circuit is configured to: receiving first configuration information; and A first portion of a second storage medium is configured based on the first configuration information.
7. The storage device according to claim 6, in, The at least one control circuit is configured to: receiving second configuration information; and A second portion of the second storage medium is configured based on the second configuration information.
8. The storage device according to claim 1, in, The at least one control circuit is configured to: receiving a request to perform an operation associated with the portion of the first memory medium; and Based on the request, the operation is performed using the portion of the second memory medium.
9. The storage device according to claim 8, in, The operation includes copying at least a portion of the data from the portion of the second storage medium to the portion of the first storage medium.
10. The storage device according to claim 8, in, The operation includes modifying a state of at least a portion of the portion of the second memory medium.
11. The storage device according to claim 1, in, The at least one control circuit is configured to: receiving a request to access the data from the portion of the first storage medium; and An access operation for the request is performed based on cache hit information corresponding to the portion of the second memory medium.
12. The storage device according to claim 11, in, Cache hit information is determined based on the placement information of the data and / or cache configuration information corresponding to the portion of the second memory medium.
13. The storage device according to claim 11, in, The at least one control circuit is configured to: In response to the cache hit information comprising a possible cache hit rate that is greater than or equal to a threshold, accessing the data from the portion of the second memory medium; and In response to a cache miss occurring when accessing the data from the portion of the second memory medium, the data is accessed from the portion of the first memory medium.
14. The memory device according to claim 11, wherein the at least one control circuit is configured to: In response to the cache hit information comprising a possible cache hit rate that is less than a threshold, the data is accessed from the portion of the first storage medium and the data is accessed from the portion of the second storage medium in parallel.
15. A storage device, include: a first storage medium; a second storage medium; as well as At least one control circuit configured to: receiving first configuration information for configuring a portion of a first storage medium corresponding to placement information of the data, and configuring the portion of the first storage medium based on the first configuration information; receiving second configuration information for configuring a portion of a second storage medium corresponding to the placement information of the data, and configuring the portion of the second storage medium based on the second configuration information; receiving a request to access the data from the portion of the first storage medium; and The data is accessed from the portion of the second storage medium based on the request.
16. The storage device according to claim 15, in, The placement information of the data is first placement information of first data, the portion of the first memory medium is a first portion of the first memory medium, the portion of the second memory medium is a first portion of the second memory medium, and the at least one control circuit is configured to: receiving third configuration information for configuring a second portion of the first storage medium corresponding to second placement information of second data, and configuring the second portion of the first storage medium based on the third configuration information; and Fourth configuration information for configuring a second portion of the second memory medium corresponding to second placement information of the second data is received, and the second portion of the second memory medium is configured based on the fourth configuration information.
17. The storage device according to claim 15, in, The at least one control circuit is configured to modify the portion of the first memory medium to an available state based on the allocation state.
18. A data processing method, include: storing data in a portion of the first storage medium based on the placement information; storing the data in a portion of a second storage medium based on the placement information; receiving a request to access the data from the portion of the first storage medium; as well as The data is accessed from the portion of the second storage medium based on the request.
19. The data processing method according to claim 18, in, The portion of the first storage medium is a first portion of the first storage medium, the portion of the second storage medium is a first portion of the second storage medium, the placement information is first placement information, the data is first data, the request is a first request, and the data processing method further includes: storing second data in a second portion of the first storage medium based on the second placement information; storing second data in a second portion of a second storage medium based on the second placement information; receiving a second request to access second data from a second portion of the first storage medium; and Second data is accessed from a second portion of a second storage medium based on the second request.
20. The data processing method according to claim 19, in: The first portion of the second memory medium operates based on the first configuration information; and The second portion of the second memory medium operates based on the second configuration information.
