Abstract: A memory system and method are provided for performing garbage collection on blocks based on their obsolescence patterns. In one embodiment, a controller of a memory system classifies each of the plurality of blocks based on its obsolescence pattern and performs garbage collection only on blocks classified with similar obsolescence patterns. Other embodiments are possible, and each of the embodiments can be used alone or together in combination.

Abstract: Data transfer between a data storage device and a peripheral device bypasses an application processor that is coupled to the data storage device and to the peripheral device. In one embodiment, the data storage device includes a memory controller configured to receive, from an application processor, a message indicating a set of logical addresses and a data transfer identifier corresponding to the set of logical addresses. The memory controller is responsive to a request for memory access that includes the data transfer identifier and that is received from a peripheral device. The memory controller is configured to respond to the request by performing a memory access operation based on the set of logical addresses.

Abstract: A method and system are disclosed for allowing access to processing resources of one or more idle memory devices to an active memory device is disclosed, where the idle and active memory devices are associated with a common host. The resources shared may be processing power, for example in the form of using a processor of an idle memory to handle some of the logical-to-physical mapping associated with a host command, or may be other resources such as RAM sharing so that a first memory has expanded RAM capacity. The method may include exchanging tokens with resource sharing abilities, operation codes and associated data relevant to the requested resources.

Abstract: A memory system and method for improving read latency of a high-priority partition are provided. In one embodiment, a memory system receives a command to store data in the memory. The memory system determines if the command specified that the data is to be stored in a standard partition in the memory or in a high-priority partition in the memory. If the command specified that the data is to be stored in a standard partition in the memory, the memory system stores the data using a first write technique. If the command specified that the data is to be stored in a high-priority partition in the memory, the memory system stores the data using a second write technique, wherein the second write technique provides improved read latency of the stored data. Other embodiments are disclosed.

Abstract: Systems and methods for configuring, controlling and operating a non-volatile cache are disclosed. A host system may poll a memory system as to the memory system's configuration of its non-volatile cache. Further, the host system may configure the non-volatile cache on the memory system, such as the size of the non-volatile cache and the type of programming for the non-volatile cache (e.g., whether the non-volatile cache is programmed according to SLC or the type of TRIM used to program cells in the non-volatile cache). Moreover, responsive to a command from the host to size the non-volatile cache, the memory system may over or under provision the cache. Further, the host may control operation of the non-volatile cache, such as by sending selective flush commands.

Abstract: A method includes, in a data storage device including a non-volatile memory, receiving, from a host device, a read command to read data from the non-volatile memory. The host device is coupled to the data storage device and includes a memory having a unified memory (UM) area. The method also includes, responsive to the read command, sending a UM read command to the host device. The UM read command instructs the host device to read the data from a location of the UM area.

Abstract: A data storage device may be configured to direct access to at least a portion of a host memory of a host device. For example, the data storage device may store data at the host memory, such as data predicted to be subject to a read request from the host device. When the data storage device receives a read request from the host device to read the data, the data storage device may send an indication to the host device to enable the host device to read the data directly from the host memory.

Abstract: Enhanced configuration of security and access control for data in a storage device is disclosed. A request is received to access an addressable memory location in a storage media within the storage device. A set of addressable memory locations with contiguous addresses identified by an address range is associated with first and second characteristics. The first characteristic is applied if the addressable memory location is within the set of addressable memory locations, and an entity is currently authenticated to and authorized to access the set of addressable memory locations. The second characteristic is applied if the addressable memory location is within the set of addressable memory locations, and no entity is currently authenticated to and authorized to access the set of addressable memory locations. The set of addressable memory locations can also be a logical partition, where the first and second characteristics are stored in a logical partition table.

Abstract: A method and system for accessing enhanced functionality on a storage device is disclosed. A hijack command is sent to the storage device that includes an identifier (such as a signature or an address). The storage device determines whether to hijack one or more subsequently commands by analyzing the subsequently commands using the identifier. For example, the storage device may analyze the subsequently received commands to determine whether the signature is in the payload of the subsequently received commands. As another example, the storage device may compare the address in the subsequently received commands with the address in the hijack command to determine whether to hijack the subsequently received commands.

Abstract: A third party facilitates preparation of a backup SSD for backing up a source SSD. Digital data of the source SSD, which includes protected and sensitive data and information, is copied to the backup SSD either by and via the third party or directly from the source SSD but under supervision of the third party. The digital data of the source SSD is copied to the backup SSD under stringent rules and only if each party (i.e., the source SSD, destination SSD, and third party) proves to a counterpart device with which it operates that it is authorized to send to it digital data or to receive therefrom digital data, depending on the device with which that party operates.

Abstract: A storage module and method for re-enabling preloading of data in the storage module are disclosed. In one embodiment, a storage module is provided with a memory and a register. In response to receiving a register-setting command, the storage module sets a value in the register to enable preloading of data in the memory. The storage module then receives the data for storage in the memory. After the storage module has determined that all of the data has been received, the storage module changes the value in the register to disable further preloading of data. In response to receiving a register-resetting command, the storage module resets the value in the register to re-enable preloading of data even though the storage module already changed the value in the register to disable further preloading of data.

Abstract: A method and system is disclosed that remaps logical block addresses (LBAs) for defragmentation that is managed at the storage device level. The remapping may include sequentially remapping LBAs where individual files are remapped so that each file is referenced by sequential LBAs. The remapping of LBAs may be performed without changes to the physical location of data.

Abstract: In one embodiment, a host device creates a super-distribution token by encrypting a content encryption key with a super-distribution key and stores the super-distribution token and encrypted content retrieved from a source storage device in a target storage device. In another embodiment, a host device provides a super-distribution token to a server, wherein the server is configured to generate an activation token from the super-distribution token, receive the activation token from the server, retrieve a content encryption key from the activation token, and decrypt encrypted content received from a storage device using the content encryption key retrieved from the activation token.

Abstract: A method and system for accessing enhanced functionality on a storage device is disclosed. A hijack command is sent to the storage device that includes an identifier (such as a signature or an address). The storage device determines whether to hijack one or more subsequently commands by analyzing the subsequently commands using the identifier. For example, the storage device may analyze the subsequently received commands to determine whether the signature is in the payload of the subsequently received commands. As another example, the storage device may compare the address in the subsequently received commands with the address in the hijack command to determine whether to hijack the subsequently received commands.

Abstract: A third party is configured to establish a virtual secure channel between a source SSD and a destination SSD via which the third party reads protected digital data from the source SSD and writes the protected digital data into the destination SSD after determining that each party satisfies eligibility prerequisites. An SSD is configured to operate as a source SSD, from which protected data can be copied to a destination SSD, and also as a destination SSD, to which protected data of a source SSD can be copied.

Abstract: A method includes, in a data storage device including a non-volatile memory, performing receiving, from a host device in a master-slave configuration with the data storage device, a first notification corresponding to a first read command. The method also includes storing, based on the first notification, a first entry in a notification queue. The first entry corresponds to the first read command. The method further includes storing first data corresponding to a second command at a location of the non-volatile memory. The location corresponds to an address to be read upon execution of the first read command. The second command is associated with an operation to be performed by the host device. The method includes, after storing the second data, setting an indicator corresponding to the first entry. The set indicator conveys that the data storage device is ready to execute the first read command.

Abstract: A method for using time from a trusted host device is disclosed. In one embodiment, an application on a memory device receives a request to perform a time-based operation from an entity authenticated by the memory device, wherein the entity is running on a host device. The application selects time from the host device instead of time from a time module on the memory device to perform the time-based operation and uses the time from the host device to perform the time-based operation. Other embodiments are disclosed, and each of the embodiments can be used alone or together in combination.

Abstract: A storage device and method for storage device state recovery are provided. In one embodiment, a storage device commences an authentication process to authenticate a host device. The authentication process comprises a plurality of phases, and the storage device stores the state of the authentication process, wherein the state indicates the phase(s) of the authentication process that have been successfully completed. After a power loss, the storage device retrieves the state of the authentication process and resumes an operation with the host device without re-performing the phase(s) of the authentication process that have been completed.

Abstract: A method and system is disclosed that remaps logical block addresses (LBAs) for defragmentation that is managed at the storage device level. The remapping may include sequentially remapping LBAs where individual files are remapped so that each file is referenced by sequential LBAs. The remapping of LBAs may be performed without changes to the physical location of data.

Abstract: A method and system defragments data during garbage collection. Garbage collection may be more efficient when the valid data that is aggregated together is related or logically linked. In particular, data from the same file or that is statistically correlated may be combined in the same blocks during garbage collection.