Patent Application
20220261352
The present disclosure relates generally to the field of storage, and more particularly to storage devices and systems having cache capability.
With advances in computing and storage systems, data storage devices may need to support different usage schemes to be effective. For instance, users may have unique performance requirements and the storage device may need to support these in real time.
Current solid-state storage devices (SSD) provide a single level of data access, and a homogeneous view of memory and data storage. With just a single level of data access to the user, the user has no way to manage the data so as to have different access performance for different types of data. In a computing system, data is categorized as non-volatile and volatile data. Non-volatile data resides in non-volatile memories in permanent data storage devices, such as SSDs or other non-volatile medium. On the other hand, volatile data resides in volatile memories, such as SRAM or DRAM. During operations, non-volatile data is often copied to the volatile memory to provide the computer processor a much faster access to the data. The data in the volatile memory is a copy of the original data in the non-volatile memory of the permanent data storage device. The copy of the data in volatile memory is known as “cached data” and the volatile memory is referred to as the “cache memory.” When the copy of the data in the cache memory is modified (e.g., by the computer processor), the original data in the permanent storage (e.g., SSD) will be updated with the modified data copy from the cache memory. The cache memory does not add to the storage capacity of the permanent storage memory (e.g., SSD) because it holds duplicate copies of data stored in the permanent storage memory.
In certain aspects of the present disclosure, a solid-state storage device (SSD) is provided that includes a controller; and non-volatile memory coupled to the controller. The non-volatile memory includes: a first portion of the non-volatile memory configured as a main memory; and a second portion of the non-volatile memory configured as a first cache memory. The controller is configured such that when data stored in the main memory is written to the first cache memory for processing, the data in the main memory is erased or otherwise considered erased and available to be rewritten with new data.
In certain aspects, a storage system is provided and includes: a host system; a first solid-state storage device (SSD) communicatively coupled to the host system; and a second SSD communicatively coupled to the host system. The first SSD includes: a first controller; and first non-volatile memory coupled to the first controller, wherein at least a portion of the first non-volatile memory is configured as a first main memory. The second SSD includes: a second controller; and second non-volatile memory coupled to the second controller. At least a portion of the second non-volatile memory is configured as a first cache memory. The first and second controllers are configured such that when data stored in the main memory is written to the first cache memory for processing, the data in the first main memory is erased or otherwise considered erased and available to be rewritten with new data.
In certain aspects, a method is provided and includes receiving, by a first solid-state storage device (SSD), a first command to write data to memory. The first SSD includes: a first controller; and first non-volatile memory coupled to the first controller. At least a portion of the first non-volatile memory is configured as a first cache memory. The method further includes: determining, by the first controller, that the data is stored in a main memory and is to be written to the first cache memory for processing; writing the data to the first cache memory; and erasing the data from the main memory or otherwise considering the data erased and available to be rewritten with new data.
For a better understanding of at least an embodiment, reference will be made to the following Detailed Description, which is to be read in conjunction with the accompanying drawings, wherein:
Before aspects of the present disclosure are described below with reference to the drawings in the description, common features may be designated by common reference numbers. Although certain examples are described herein with reference to a data storage system, it should be appreciated that techniques described herein are applicable to other implementations. Further, it is to be appreciated that certain ordinal terms (e.g., “first” or “second”) may be provided for ease of reference and do not necessarily imply physical characteristics or ordering. Therefore, as used herein, an ordinal term (e.g., “first,” “second,” “third,” etc.) used to modify an element, such as a structure, a component, an operation, etc., does not necessarily indicate priority or order of the element with respect to another element, but rather distinguishes the element from another element having a same name (but for use of the ordinal term). In addition, as used herein, indefinite articles (“a” and “an”) may indicate “one or more” rather than “one.” Further, an operation performed “based on” a condition or event may also be performed based on one or more conditions, or events not explicitly recited. As used herein, “exemplary” may indicate an example, an implementation, and/or an aspect, and should not be construed as limiting or as indicating a preference or a preferred example, implementation, and/or aspect.
As stated above, non-volatile data is traditionally often copied to the volatile memory to provide a computer processor with much faster access to the data. In such traditional systems, the cache memory does not add to the storage capacity of the permanent storage memory (e.g., the SSD). In certain aspects, the present disclosure provides devices, systems, and methods that enable cache memory to be part of the permanent storage (SSD), and in some embodiments have the cache memory add to the capacity of the permanent storage.
In certain embodiments, a computing system may include a server and one or more storage devices (or subsystems). A storage device is typically composed of a solid-state drive (or device) (SSD). In certain aspects, the present disclosure provides devices, systems, and methods related to one or more storage devices, such as one or more SSDs, that can be customized to support specialized embedded (or integrated) caching mechanisms based on user needs. In certain aspects, the present disclosure provides for an SSD architecture that supports multiple embedded cache mechanisms aimed for various performance requirements, such as latency, capacity, IOPS, and media type. It should be appreciated that reference to an SSD is used broadly herein and can refer to a stand-alone SSD device or embedded SSD device.
In certain aspects, the present disclosure describes devices, systems, and methods of accessing stored data in a storage device using specific access modes to achieve the desired read/write performance. This can be used in programming data to, or reading data from, a storage system, a memory card, or a set of memory cards, according to the use of an architecture using a centralized or distributed controller scheme for performing the function. For example, a controller of a data storage device may receive a data packet to be stored in a number of memory devices. The controller may partition the data word into multiple portions and the multiple portions may be interleaved across multiple memory devices. Also, the controller may simultaneously start parallel reads from several memory devices. Additionally, prior to the multiple data packets being stored into, or read from, the memory card, or storage system, a logical address corresponding to the data packets may be mapped to multiple physical addresses that indicate where each of the multiple data packets is to be stored in, or read from, the memory card or storage system. The writing of data and reading of data can be performed simultaneously across several memory cards in the storage system using a distributed control architecture. In addition, the storage device may be configured as to have storage areas, blocks, and memory partitions, to operate with different performance so as to increase the performance of the data access. These various configured partitions can be seen as providing “cache” capabilities internal to the storage device or SSD.
The subject matter of the present disclosure may be applicable to a wide range of areas, including, but not limited to, storage, networking, communication and computing systems, appliances on the Internet, remotely controlled appliances, and design of reliable systems. In certain aspects, a computing system includes a server and a storage subsystem. The storage subsystem can be typically composed of one or more solid-state drives (SSDs). In certain embodiments, the present disclosure provides devices, systems, and methods related to one or more storage devices, such as one or more SSDs, that can be customized to support specialized embedded caching mechanisms based on user needs. In certain aspects, the present disclosure provides for a SSD architecture that supports the multiple embedded cache mechanisms aimed for various performance requirements, such as latency, capacity, TOPS, and media type.
The memory card 160 includes an array of memory devices, of which a memory device 170 is a representative device. For the sake of clarity and brevity, all of the memory devices may also be collectively referred to herein as “the memory devices 170.” The memory card 160 may include a large number of memory devices 170. The memory devices 170 may include, for example, Flash, ReRam, MRAM, any other non-volatile memory (NVM) technology, or a combination of different technology devices. In certain embodiments, the memory card 160 can include a set of the memory devices 170 configured to perform as high performance memory devices, and thus provide an internal “cache” memory capability. For example, during the initialization of the system, a set of the memory devices can be designated as being used for caching purposes. The high performance cache memory can be utilized for storing temporary data, or as a staging place for data from the main memory, for instance.
The discrete components 120 may be implemented to assist with the operation of the storage device 100, and may include passive components such as capacitors, resistors, and inductors, as well as active components such as diodes and transistors. This list of components is an illustrative list and not an exhaustive list.
In certain embodiments, the storage device 100 may operate in data centers where usage encompasses several scenarios. Some of the data may be accessed frequently and is “hot”, while other data may be accessed infrequently and is practically “cold”. Therefore, “hybrid” requirements may be present for the storage device 100 to accommodate (or account for). To accommodate the hybrid requirements, storage device 100 supports technology that can be tailored to different usage scenarios. In alternative embodiments, instead of having a centralized control strategy, which would not adequately accommodate the hybrid requirements, the storage device 100 may include distributed data control, while having the controller 130 function as the main controller that manages host accesses, memory management, and other background tasks.
The controller 130 shown in
Each memory bank 231, 232, 233, 234, 235, 236, and 237 can have one or more memory devices. The memory banks 230 and 231 are shown having memory devices 241; the memory banks 232 and 233 are shown having memory devices 242; the memory banks 234 and 235 are shown having memory devices 244; and the memory banks 236 and 237 are shown having memory devices 243. The memory devices 241, 242, 243, and 244 shown are exemplary and are not an exhaustive list. Each memory bank, such as memory bank 230, can have several memory devices, and can have a different number of memory devices than shown in the example embodiment of
The memory modules 270, 271, 272 and 273 can be mounted directly on the main board 201 of
The main memory 330 includes non-volatile memory, which can include Flash, ReRam, MRAM, any other non-volatile memory (NVM) technology, or a combination of different technology devices. The storage device 310 also includes a management structure, such as a controller, and may also include peripheral supporting devices (not shown in
The partition of the non-volatile memory 410 that has been designated as the cache memory 430 has its memory components configured to perform at higher rates when compared with the regular non-volatile memory of the main memory 420. The cache memory 430 can have beneficial attributes, such as low latency, high IOPS (TO operations per second), capacity, and media type (e.g., thumbdrives, compact flash, SD cards, etc.). This is not an exhaustive list of attributes but serves as examples. In one embodiment, the storage device 400 includes (or corresponds to) an SSD.
The storage device 400 is shown communicatively coupled to the host 450 via a bus 440. The host 450 communicates with the storage device 400 (e.g., the main memory 420 and the cache memory 430) via the bus 440. In an embodiment, the host is a compute server. The bus 440 may include, for example, a PCIe, SATA, USB, Open CAPI, or any other standard or proprietary interface.
In an embodiment, the storage device 400 can include the centralized control architecture of the storage device 100 of
In another embodiment, the storage device 400 can include the distributed control architecture of the storage device 200 of
The main memory partition and the cache memory partition can be formed in a variety of manners as desired. For example, in one embodiment, the main memory 420 and the cache memory 430 can be partitioned on a per memory device basis such that each of the memory devices 241 through 244 can be configured to work as the main memory 420 or the cache memory 430. In this way, for instance, the attributes of the memory devices 241 through 244 can be utilized to determine whether better suited for the main memory 420 or the cache memory 430.
In another embodiment, the main memory 420 and the cache memory 430 can be partitioned on a per memory bank basis such that each of the memory banks 230 through 237 can be configured to work as the main memory 420 or the cache memory 430. In this way, for instance, the attributes of the memory banks 230 through 237 can be utilized to determine whether better suited for the main memory 420 or the cache memory 430.
In yet another embodiment, the main memory 420 and the cache memory 430 can be partitioned on a per module basis such that all of the memory devices on a memory module are configured to work as the main memory 420 or the cache memory 430. In this way, for instance, the attributes of the memory modules 270 through 273, their respective distributed controllers 210 through 213, or both, can be utilized to determine whether better suited for the main memory 420 or the cache memory 430.
In certain aspects, when data stored in one partition is written to another partition (also referred to herein as being “moved from one partition to another partition,” such as from the main memory 420 to the cache memory 430, or vice versa), no copy of the data is retained in the originating partition. For example, the controller on the storage device can erase (or delete) the data from the originating partition or otherwise consider (or treat, function as though, etc.) the data as if it was erased by considering the memory holding the data to be available for new data to be written to) once the data is written to the target partition. In this way, there is no loss in the storage capacity since there are not two duplicate copies of the data stored.
In a preferred embodiment, when data is moved to the cache memory 430 from the main memory 420, no copy of the data is retained in the main memory 420. In this way, there is no loss in the storage capacity since there are not two duplicate copies of the data. When the processing of the data in the cache memory 430 is complete, the processed data can be moved to the main memory 420 and erased from the cache memory 430. The cache memory 430 is now available to be programmed with new data that requires manipulation by the host 450.
The cache memories 530 or 540 can be created such that the memory devices that make up the cache memories 530 and 540 are configured to provide a desired performance, such as a higher performance than provided by the main memory 520. The attributes of the memory devices can be utilized to determine whether better suited for the main memory 520 or the cache memories 530 and 540. The attributes can relate to latency, TOPS, capacity, media type, etc., for example.
The storage device 500 is shown communicatively coupled to the host 560 (e.g., a compute server) via a bus 550. The host 560 communicates with the storage device 500 (e.g., the main memory 520 and the cache memories 530 and 540) via the bus 550. The bus 550 may include, for example, a PCIe, SATA, USB, Open CAPI, or any other standard or proprietary interface.
In an embodiment, the storage device 500 can include the centralized control architecture of the storage device 100 of
In another embodiment, the storage device 500 can include the distributed control architecture of the storage device 200 of
The main memory partition 520 and the cache memory partitions 530 and 540 can be formed in a variety of manners as desired. For example, in one embodiment, the main memory 520 and the cache memories 530 and 540 can be partitioned on a per memory device basis such that each of the memory devices 241 through 244 can be configured to work as the main memory 520 or one of the cache memories 530 and 540. In this way, for instance, the attributes of the memory devices 241 through 244 can be utilized to determine whether better suited for the main memory 520 or the cache memories 530 and 540.
In another embodiment, the main memory 520 and the cache memories 530 and 540 can be configured on a per memory bank basis such that any of the memory banks 230 through 237 can be configured to work as the main memory 520 or one of the cache memories 530 and 540. In this way, for instance, the attributes of the memory banks 230 through 237 can be utilized to determine whether better suited for the main memory 520 or the cache memories 530 and 540.
In yet another embodiment, the main memory 520 and the cache memories 530 and 540 can be partitioned on a per module basis such that each of the memory modules 270 through 273 can be configured to work as the main memory 520 or one of the cache memories 530 and 540. In this way, for instance, the attributes of the memory modules 270, 271, 272, and 273, their respective distributed controllers 210, 211, 212, and 213, or both, can be utilized to determine whether better suited for the main memory 520 or the cache memories 530 and 540.
In certain aspects, when data is moved from one partition to another partition (e.g., the (e.g., from the main memory 520 to one of the cache memories 530 and 540, or vice versa), no copy of the data is retained in the originating partition. For example, the controller in the storage device can erase the data from the originating partition (or otherwise consider the data as if it was erased by considering the memory holding the data to be available for new data to be written to) once the data is written to the target partition. In this way, there is no loss in the storage capacity since there are not two duplicate copies of the data stored.
In a preferred embodiment, the storage device 500 is configured such that as data is moved between partitions of the non-volatile memory 510 (e.g., between the main memory 520, the cache memory 530, and the cache memory 540), there will be only one copy of data stored in the non-volatile memory 510. In such case, the data can be in the main memory 520 or in either of the cache memories 530 and 540 at a given time. For example, when data is moved to one of the cache memories 530 and 540 from the main memory 520, no copy of the data is retained in the main memory 520. In this way, there is no loss in the storage capacity since there are not two duplicate copies of the data stored in the non-volatile memory 510. When the processing of the data in the cache memory 530 or 540 is complete, the processed data can be moved to the main memory 520 and erased from the cache memory 530 or 540. The cache memory 530 or 540 is then available to be programmed with new data that requires manipulation by the host 560.
In certain aspects, a storage system can include multiple storage devices, with each storage device configured to work as main memory, cache memory, or combination thereof. For example, one storage device (e.g., SSD) can include non-volatile memory configured as the main memory of the storage system; another storage device (e.g., SSD) can include non-volatile memory configured as the cache memory; and yet another storage device (e.g., SSD) can include non-volatile memory configured as another cache memory. The non-volatile memory on a storage device can be partitioned as well. For example, one storage device can include main memory and one or more cache memories, such as the storage devices 400 and 500 of
The storage system 600 can include one or more communication interfaces (not shown). For example, the storage devices 610, 630, and 650 can be communicatively coupled to each other via a communication interface 680, such as an electrical bus 680. Another communication interface 690 can enable communication between a host 670 and the storage devices 610, 630, and 650, and may include, for example, a PCIe, SATA, USB, Open CAPI, or any other standard or proprietary interface.
In an embodiment, the storage devices 610, 630, and 650 can include the centralized control architecture of the storage device 100 of
In an embodiment, one or more of the storage devices 610, 630, and 650 can include an SSD and can be part of a storage appliance having one or more SSDs. For example,
At block 710 of a method 700, a storage device (or system) receives a read or write command from a host. At block 720, the command is decoded to determine if the command is a read or write command. If the command is determined to be a read command, then a search is performed on the physical address for the data to be read, as represented by block 724. At block 726, a determination is made as to identify which partition (e.g., the main memory or one of the cache memories of
In the example data storage device 100 with centralized controller shown in
In the example data storage device 200 with distributed controllers in
If at block 720, the command is determined to be a write command, then a search is performed at block 730 to determine and select which partition the data is to be written. For example, if the write command is to store the data, then the search is performed to determine and select the appropriate partition configured as the main memory. On the other hand, if the write command is to enable processing of the data (e.g., as indicated in the instructions (or command) by the host), then the search is performed to determine and select the appropriate partition configured as the cache memory. If more than one cache memory partition exists, one can be selected based on various factors, such as which partition has available memory space, or which partition meets the appropriate performance requirements (e.g., latency, capacity, IOPS, media type, etc.), etc. This attribute information can be maintained, for example, in the page table along with the physical addresses and corresponding partitions. At block 740, the data is written to the selected partition. If the data written to the selected partition existed in another partition, then at block 750, the data in the other partition is not maintained (e.g., deleted, erased, or otherwise considered as if erased). For example, if data is transferred from the main memory to one of the cache memories, then the data in the main memory will be erased once the data is written in the cache memory. In this way, there are not duplicate copies of the written data, which would reduce storage capacity. In an embodiment, the operations of blocks 730, 740 and 750 can be performed by a controller on the storage device, such as the centralized controller 130 of
For example, in the data storage device 100 with centralized controller shown in
If, on the other hand, the data is to be further processed, then the data can be written to (block 740) a cache memory (e.g., cache memories 430, 530, 540, 640, or 660) having higher performance attributes than the main memory. Once written to the cache memory, the data can be processed accordingly. If the data was originally stored in the main memory, then the controller 130 can erase (block 750) the copy of the data in the main memory once the data is written to the cache memory. In some instances, the processed data can be moved back to the main memory when processing is complete. In such case, for example, the processed data can be written to the main memory and the copy of the processed data in the cache memory can be erased.
In the example data storage device 200 with distributed controllers in
For instance, the partition can be selected based on the function to be performed on the data and the most appropriate performance attributes to perform the function (block 730). For example, if the data is to be stored, then the data can be sent to the appropriate memory module (270, 271, 272, or 273) having memory devices configured to work as the main memory (e.g., the main memory 420, 520, or 620). On the other hand, if the data is to be further processed, then the data can be sent to the appropriate memory module (270, 271, 272, or 273) with memory devices configured as the cache memory (e.g., cache memories 430, 530, 540, 640, or 660) having higher performance attributes than the main memory. The respective distributed controller (e.g., 210, 211, 212, or 213) controlling the appropriate memory module can then write the data to a physical address corresponding to the selected cache memory (block 740). If a copy of the data was stored in another partition, then the copy of the data can be erased from the other partition (block 750).
Once written to the cache memory, the data can be processed accordingly. In some instances, the processed data can be moved back to the main memory when processing is complete. In such case, for example, the main controller 280 can instruct a distributed controller to write the processed data to a memory module configured to work (at least partially) as the main memory.
The process may be repeated for subsequent commands received from the host, as represented by the arrows from blocks 717 and 750 back to the block 710. It should be appreciated that in some embodiments, multiple commands can be processed simultaneously by one or more controllers.
It should be appreciated that the data to be read can be stored at more than one physical address, and that the blocks 710 through 728 can also apply to read operations for data stored at more than one physical address. For example, if the data to be read is located at multiple physical addresses, then the search is performed on the physical addresses (block 724), the one or more corresponding partitions are identified, and the data at the physical addresses on the identified partitions is read (block 726). As for write operations, one or more partitions can be determined and selected (block 730) and the data written to multiple physical addresses in one or more selected partitions (block 740).
As shown, the host system 800 includes a system bus 802, which is coupled to a microprocessor 803, a Read-Only Memory (ROM) 807, a volatile Random Access Memory (RAM) 805, as well as other nonvolatile memory 806. In the illustrated embodiment, microprocessor 803 is coupled to cache memory 804. A system bus 802 can be adapted to interconnect these various components together and also interconnect components 803, 807, 805, and 806 to other devices, such as a display controller and display device 808, and to peripheral devices such as input/output (“I/O”) devices 810. Types of I/O devices can include keyboards, modems, network interfaces, printers, scanners, video cameras, or other devices well known in the art. Typically, I/O devices 810 are coupled to the system bus 802 through I/O controllers 809. In one embodiment the I/O controller 809 includes a Universal Serial Bus (“USB”) adapter for controlling USB peripherals or other type of bus adapter.
RAM 805 can be implemented as dynamic RAM (“DRAM”), which requires power continually in order to refresh or maintain the data in the memory. The other nonvolatile memory 806 can include a magnetic hard drive, magnetic optical drive, optical drive, DVD RAM, solid-state storage drive, or other type of memory system that maintains data after power is removed from the system. While
In certain aspects of the present disclosure, a general implementation of a solid-state storage device (SSD) includes a controller; and non-volatile memory coupled to the controller. The non-volatile memory includes: a first portion of the non-volatile memory configured as a main memory; and a second portion of the non-volatile memory configured as a first cache memory. The controller is configured such that when data stored in the main memory is written to the first cache memory for processing, the data in the main memory is erased or otherwise considered erased and available to be rewritten with new data.
In certain embodiments, the controller is configured such that when the processed data is written to the main memory, the processed data in the first cache memory is erased or otherwise considered erased and available to be rewritten with new data.
In certain embodiments, the non-volatile memory further includes a third portion of the non-volatile memory configured as a second cache memory. The controller is configured such that when data stored in the main memory is written to the second cache memory for processing, the data in the main memory is erased or otherwise considered erased and available to be rewritten with new data. The controller is configured such that when the processed data is written to the main memory, the processed data in the second cache memory is erased or otherwise considered erased and available to be rewritten with new data.
In certain embodiments, the second portion of the non-volatile memory includes at least one memory device consisting of the group: Flash, resistive random access memory (ReRAM), and magnetoresistive random access memory (MRAM). The at least one memory device is configured to have higher data access performance than the first portion of the non-volatile memory.
In certain embodiments, the controller is configured as a centralized controller coupled to one or more memory cards having the non-volatile memory disposed thereon.
In certain embodiments, the controller is configured as a main controller. The SSD further includes one or more distributed controllers coupled to the main controller and the non-volatile memory. In certain embodiments, the one or more distributed controllers include: a first distributed controller configured to control the main memory; and a second distributed controller configured to control the first cache memory.
In certain aspects, another general implementation of a storage system includes: a host system; a first solid-state storage device (SSD) communicatively coupled to the host system; and a second SSD communicatively coupled to the host system. The first SSD includes: a first controller; and first non-volatile memory coupled to the first controller, wherein at least a portion of the first non-volatile memory is configured as a first main memory. The second SSD includes: a second controller; and second non-volatile memory coupled to the second controller. At least a portion of the second non-volatile memory is configured as a first cache memory. The first and second controllers are configured such that when data stored in the main memory is written to the first cache memory for processing, the data in the first main memory is erased or otherwise considered erased and available to be rewritten with new data.
In certain embodiments, the controller is configured such that when the processed data is written to the first main memory, the processed data in the first cache memory is erased or otherwise considered erased and available to be rewritten with new data.
In certain embodiments, the first non-volatile memory includes a second portion of the first non-volatile memory configured as a second cache memory.
In certain embodiments, the second non-volatile memory includes a second portion of the second non-volatile memory configured as a second main memory.
In certain embodiments, the second non-volatile memory includes a second portion of the second non-volatile memory configured as a second cache memory.
In certain embodiments, the storage system further includes a third SSD communicatively coupled to the host system. The third SSD includes: a third controller; and a third non-volatile memory coupled to the third controller, wherein at least a portion of the third non-volatile memory is configured as a second cache memory. The first and third controllers are configured such that when data stored in the main memory is written to the second cache memory for processing, the data in the first main memory is erased or otherwise considered erased and available to be rewritten with new data. The first and third controllers are configured such that when the processed data is written to the first main memory, the processed data in the second cache memory is erased or otherwise considered erased and available to be rewritten with new data.
In certain embodiments, the portion of the second non-volatile memory that is configured as the first cache memory comprises at least one memory device consisting of the group: Flash, resistive random access memory (ReRAM), and magnetoresistive random access memory (MRAM); and wherein the at least one memory device is configured to have higher data access performance than the portion of the first non-volatile memory that is configured as the main memory.
In certain embodiments, at least one of the first and second controllers is configured as a centralized controller coupled to one or more memory cards having the respective first or second non-volatile memory disposed thereon.
In certain embodiments, either the first or second controller is configured as a main controller, and wherein the respective first or second SSD further comprises one or more distributed controllers coupled to the main controller and the respective first or second non-volatile memory. In certain embodiments, the first non-volatile memory comprises a second portion of the first non-volatile memory configured as a second cache memory. The first controller is configured as a main controller. The first SSD further includes: a first distributed controller coupled to the main controller and the first non-volatile memory; and a second distributed controller coupled to the main controller and the first non-volatile memory. The first distributed controller is configured to control the main memory. The second distributed controller is configured to control the second cache memory.
In certain aspects, another general implementation of a method includes receiving, by a first solid-state storage device (SSD), a first command to write data to memory. The first SSD includes: a first controller; and first non-volatile memory coupled to the first controller. At least a portion of the first non-volatile memory is configured as a first cache memory. The method further includes: determining, by the first controller, that the data is stored in a main memory and is to be written to the first cache memory for processing; writing the data to the first cache memory; and erasing the data from the main memory or otherwise considering the data erased and available to be rewritten with new data.
In certain embodiments, the method further includes, after the processing of the data, writing the processed data to the main memory and erasing the data from the first cache memory or otherwise considering the data erased and available to be rewritten with new data.
In certain embodiments, the first non-volatile memory further includes a second portion of the first non-volatile memory configured as the main memory. The first controller performs: the writing of the data to the first cache memory; the erasing of the data from the main memory or the otherwise considering of the data as erased and available to be rewritten with new data; the writing of the processed data to the main memory; and the erasing of the data from the first cache memory or the otherwise considering of the data as erased and available to be rewritten with new data.
In certain embodiments, the portion of the first non-volatile memory that is configured as the first cache memory comprises at least one memory device consisting of the group: Flash, resistive random access memory (ReRAM), and magnetoresistive random access memory (MRAM); and wherein the at least one memory device is configured to have higher data access performance than the second portion of the first non-volatile memory that is configured as the main memory.
In certain embodiments, the method further includes: receiving, by the first SSD, a second command to read data from memory; determining, by the first controller, that the data is stored at a physical address located on either the main memory or the first cache memory; and reading the data from the physical address located on the respective main memory or first cache memory.
In certain embodiments, the first controller is configured as a main controller. The first SSD further includes a first distributed controller coupled to the main controller and the first non-volatile memory. The first distributed controller is configured to control the main memory. The first distributed controller performs: the erasing of the data from the main memory or the otherwise considering of the data as erased and available to be rewritten with new data; and the writing of the processed data to the main memory; and a second distributed controller coupled to the main controller and the first non-volatile memory. The second distributed controller configured to control the first cache memory. The second distributed controller performs: the writing of the data to the first cache memory; and the erasing of the data from the first cache memory or the otherwise considering of the data as erased and available to be rewritten with new data.
In certain embodiments, a second SSD is coupled to the first SSD. The second SSD includes: a second controller; and second non-volatile memory coupled to the first controller. At least a portion of the second non-volatile memory is configured as the main memory. The first controller performs: the writing of the data to the first cache memory; and the erasing of the data from the first cache memory or the otherwise considering of the data as erased and available to be rewritten with new data. The second controller performs: the erasing of the data from the main memory or the otherwise considering of the data as erased and available to be rewritten with new data; and the writing of the processed data to the main memory.
Throughout the foregoing description, and for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the described techniques. It will be apparent, however, to one skilled in the art that these techniques can be practiced without some of these specific details. Although various embodiments that incorporate these teachings have been shown and described in detail, those skilled in the art could readily devise many other varied embodiments or mechanisms to incorporate these techniques. Also, embodiments can include various operations as set forth above, fewer operations, or more operations, or operations in an order. Accordingly, the scope and spirit of the invention should only be judged in terms of any accompanying claims that may be appended, as well as any legal equivalents thereof.
Reference throughout the specification to “one embodiment” or “an embodiment” is used to mean that a particular feature, structure or characteristic described in connection with an embodiment is included in at least one embodiment. Thus, the appearance of the expressions “in one embodiment” or “in an embodiment” in various places throughout the specification are not necessarily referring to the same embodiment. Further, the particular features, structures or characteristics may be combined in any suitable manner in one or several embodiments. Although the present invention has been described above with reference to specific embodiments, it is not intended to be limited to the specific form set forth herein. Rather, embodiments other than those specific described above are equally possible within the scope of any accompanying claims. Moreover, it should be appreciated that the terms “comprise/comprises” or “include/includes”, as used herein, do not exclude the presence of other elements or steps. Furthermore, although individual features may be included in different claims, these may possibly advantageously be combined, and the inclusion of different claims does not imply that a combination of features is not feasible and/or advantageous. In addition, singular references do not exclude a plurality. Finally, reference signs in the claims are provided merely as a clarifying example and should not be construed as limiting the scope of the claims in any way.
For purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the description. It should be apparent, however, to one skilled in the art that embodiments of the disclosure can be practiced without these specific details. In some instances, modules, structures, processes, features, and devices are shown in block diagram form in order to avoid obscuring the description. In other instances, functional block diagrams and flow diagrams are shown to represent data and logic flows. The components of block diagrams and flow diagrams (e.g., modules, blocks, structures, devices, features, etc.) may be variously combined, separated, removed, reordered, and replaced in a manner other than as expressly described and depicted herein. It should be appreciated that the block diagrams may include additional components that are not necessarily shown or described, but which have been left out for the sake of clarity and brevity.
Various components and modules described herein may include software, hardware, or a combination of software and hardware. The components and modules may be implemented as software modules, hardware modules, special-purpose hardware (e.g., application specific hardware, ASICs, DSPs, etc.), embedded controllers, hardwired circuitry, hardware logic, etc. Software content (e.g., data, instructions, and configuration) may be provided via an article of manufacture including a non-transitory, tangible computer or machine-readable storage medium, which provides content that represents instructions that can be executed. The content may result in a computer performing various functions/operations described herein.
In general, the processes and features described herein may be implemented as part of an operating system or a specific application, component, program, object, module, or series of instructions referred to as “programs”. For example, one or more programs may be used to execute specific processes described herein. The programs typically comprise one or more instructions in various memory that, when read and executed by a processor, cause the processor to perform operations to execute the processes and features described herein. The processes and features described herein may be implemented in software, firmware, hardware (e.g., an application specific integrated circuit, or a field-programmable gate array (FPGA)), or any combination thereof. For example, the controllers described herein can include one or more processors (or processing units) that may be implemented as described above to execute the instructions. The term “processor” is used broadly herein and may include one or more processing units or circuitry, such as one or more embedded or non-embedded processors, microprocessors, hard and soft microprocessor cores, etc.
In an implementation, the processes and features described herein may be implemented as a series of executable modules run by a processor (e.g., in a computer system, individually, collectively in a distributed computing environment, embedded in a controller, etc.). The foregoing modules may be realized by hardware, executable modules stored on a computer-readable medium (or machine-readable medium), or a combination of both. For example, the modules may comprise a plurality or series of instructions to be executed by a processor in a hardware system. Initially, the series of instructions may be stored in memory, such as on a storage device. However, the series of instructions can be stored on any suitable computer readable storage medium. Furthermore, the series of instructions need not be stored locally, and could be received from a remote storage device, such as a server on a network, via the network interface. In various implementations, a module or modules can be executed by a processor or multiple processors in one or multiple locations, such as multiple servers in a parallel processing environment
A computer or machine readable non-transitory storage medium includes any mechanism that provides (i.e., stores and/or transmits) information in a form accessible by a computer (e.g., computing device, electronic system, etc.), such as recordable/non-recordable media (e.g., read only memory (ROM), random access memory (RAM), magnetic disk storage media, optical storage media, flash memory devices, etc.); or any type of medium suitable for storing, encoding, or carrying a series of instructions for execution by a processor to perform any one or more of the processes and features described herein. The content may be directly executable (“object” or “executable” form), source code, or difference code (“delta” or “patch” code). A computer readable storage medium may also include a storage or database from which content can be downloaded. A computer readable medium may also include a device or product having content stored thereon at a time of sale or delivery. Thus, delivering a device with stored content, or offering content for download over a communication medium may be understood as providing an article of manufacture with such content described herein.
This application is a continuation of U.S. patent application Ser. No. 16/188,944, filed on Nov. 13, 2018, which claims the benefit of U.S. Provisional Application No. 62/584,703, filed Nov. 10, 2017, the entireties of which are all incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| 62584703 | Nov 2017 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 16188944 | Nov 2018 | US |
| Child | 17739033 | US |
