Abstract: The present disclosure generally relates to estimating when data to be written will be read or re-written prior to actually writing the data to the memory device. The estimating can be used to smartly route the data to the appropriate memory location at the writing stage or to evict the data from a hot memory location to a colder memory location. To perform the estimating, typical traces or data may be used as may the metadata of the data. Separating data according to the data “temperature” (i.e. the expected access time and frequency), and usage to optimize the SLC partition usage has meaningful impact on several storage metrics such as performance and endurance.

Abstract: A data storage device comprises a first memory, a second memory, and a controller. The first memory has a faster access time than the second memory. The controller is configured to store host-initialization code in the first memory, store a copy of the host-initialization code in the second memory, determine that the copy of the host-initialization code should be designated as the main version of the host-initialization code, and relocate the copy of the host-initialization code to the first memory, which makes the copy of the host-initialization code the main version of the host-initialization code that is accessed to boot-up the host. Other embodiments are possible, and each of the embodiments can be used alone or together in combination.

Abstract: A data storage device and method for dynamic prediction of random read with low memory consumption are provided. In one embodiment, a data storage device comprises a volatile memory, a non-volatile memory, and a controller. The controller is configured to allocate an amount of space in the volatile memory for a history pattern matcher data structure used to predict next read commands from a host to read data stored in the non-volatile memory; determine an accuracy of the predicted next read commands; and based on the determined accuracy, dynamically allocate a different amount of space in the volatile memory for the history pattern matcher data structure. Other embodiments are possible, and each of the embodiments can be used alone or together in combination.

Abstract: A data storage device includes a memory device and a controller coupled to the memory device. When a read threshold calibration operation occurs, less than all of the pages of a representative wordline is sensed, such that read thresholds of less than all of the pages of the representative wordline is obtained. The obtained read thresholds and one or more physical conditions of the representative wordline are provided to a model to obtain the other read thresholds of the remaining pages of the representative wordline that were not sensed. The model correlates read thresholds of one page to another page of the same representative wordline and accounts for the one or more physical conditions of the representative wordline.

Abstract: A data storage device and method for device-initiated hibernation are provided. In one embodiment, the data storage device comprises a non-volatile memory and a controller. The controller is configured to: receive, from a host during a set-up phase of a hibernation process, a plurality of write commands with a current state of a volatile memory in the host; store the plurality of write commands in a queue, wherein the plurality of write commands are not executed during the set-up phase of the hibernation process; receive a trigger from the host to perform an execution phase of the hibernation process; and in response to receiving the trigger, execute the plurality of write commands to store the current state of the host's volatile memory in the non-volatile memory of the data storage device. Other embodiments are possible, and each of the embodiments can be used alone or together in combination.

Abstract: A data storage device and method for prediction-based improved power-loss handling. In one embodiment, a data storage device is provided comprising a memory and a controller. The controller is configured to predict a probability of an ungraceful shutdown of the data storage device; determine whether the probability is greater than a threshold; and in response to determining that the probability is greater than the threshold, reduce a risk of data loss that would occur in response to the ungraceful shutdown of the data storage device. Other embodiments are possible, and each of the embodiments can be used alone or together in combination.

Abstract: A method and apparatus for cache management and eviction polices using unsupervised reinforcement learning schemes is disclosed.

Abstract: A data storage device and method for race-based data access in a multiple host memory buffer system are provided. In one embodiment, the data storage device stores data in a plurality of host memory buffers in the host instead of in just the host memory buffer usually associated with the data. To read the data, the data storage device sends read commands to all of the host memory buffers. That way, even if some of the host memory buffers are busy, the data can be returned from another one of the host memory buffers. In future reads in similar workloads, a read command can be sent to the host memory buffer that returned the data. Other embodiments are possible, and each of the embodiments can be used alone or together in combination.

Abstract: Systems and methods for deduplication of storage device encoded data are described. The storage device may initiate a deduplication process and determine a encoded target data block and at least one encoded comparison data block. The storage device may compare the encoded target data block to the encoded comparison data blocks to determine similarity values. Based on the similarity values, the storage device may determine duplicate data units and eliminate extra duplicate data units.

Abstract: The disclosure relates in some aspects to on-chip processing circuitry formed within the die of a non-volatile (NVM) array to perform data searches. In some aspects, the die includes components configured to sense wordlines of stored data in the NVM array by applying voltages on the wordlines serially, and then search for an input data pattern within the serially-sensed wordlines. In some examples, the components of the die include latches and circuits configured to perform bitwise latch logic search operations. In other examples, the search components are configured with under-the-array or next-to-the-array dedicated search circuitry that uses registers and/or random access memory (RAM). Other aspects relate to a separate controller device for controlling the on-chip NVM search operations. For example, the controller may determine whether to search for data using search components of the NVM die or processors of the controller based, e.g., on a degree of fragmentation of data.

Abstract: A data storage device and method for adaptive host memory buffer allocation based on virtual function prioritization are provided. In one embodiment, a data storage device is provided comprising a memory, an interface, and a controller. The controller is configured to receive priority information of each of a plurality of virtual functions in the host and allocate space in the host memory buffer for each of the plurality of virtual functions based on the priority information. The controller is further configured to dynamically reallocate the space. Other embodiments are possible, and each of the embodiments can be used alone or together in combination.

Abstract: A data storage device and method for multi-level conditional prediction of future random read commands are provided. In one embodiment, a data storage device is provided comprising a memory and a controller. The controller is configured to receive a random read command from a host, wherein the received random read command is associated with a stream; predict a next stream to be received from the host; and predict a next random read command to be received from the host based on the received random read command and the predicted next stream. Other embodiments are possible, and each of the embodiments can be used alone or together in combination.

Abstract: Recurrent Neural Networks (RNNs) wherein a non-volatile memory (NVM) array provides a memory bank for the RNN. The RNN may include a Neural Turning Machine (NTM) and the memory bank may be an NTM matrix stored in the NVM array. In some examples, a data storage device (DSD) that controls the NVM array includes both a data storage controller and a separate NTM controller. The separate NTM controller accesses the NTM matrix of the NVM array directly while bypassing flash translation layer (FTL) components of the data storage controller. Additionally, various majority wins error detection and correction procedures are described, as well as various disparity count-based procedures.

Abstract: Methods and apparatus for power management in data storage devices are provided. One such data storage device (DSD) includes a non-volatile memory (NVM), a set of hardware processing engines, and a power sensor to detect a total power consumption of the set of hardware processing engines. A processor is configured to determine a power-per-processing event value for each of the set of processing engines based on total power consumption measurements, then control delivery of power to the processing engines based on the power-per-processing event values in accordance with a power budget. In some examples, the DSD employs a least-squares procedure to estimate the power-per-processing event values so the values can be determined without needing to measure the individual power consumption of the processing engines. Exemplary processing engines include a Read engine, a Write engine, etc. A recursive least-squares update procedure is also described.

Abstract: Methods and apparatus for power management in data storage devices are provided. One such data storage device (DSD) includes a non-volatile memory (NVM), a set of hardware processing engines, and a power sensor to measure a total power consumption of the set of hardware processing engines. A processor is configured to determine a power-per-processing event value for each of the set of processing engines based on total power consumption measurements, then control delivery of power to the processing engines based on the power-per-processing event values in accordance with a power budget. In some examples, the DSD employs a least-squares procedure to estimate the power-per-processing event values so the values can be determined without needing to measure the individual power consumption of the processing engines. Procedures are also provided for assessing the accuracy of the power-per-processing event values and for controlling further operations based on the assessment.

Abstract: A data storage device and method for providing an adaptive data path are disclosed. In one embodiment, a data storage device is in communication with a host comprising a first processor (e.g., a graphics processing unit (GPU)), a second processor (e.g., a central processing unit (CPU)), and a queue. The data storage device chooses a data path to use to communicate with the queue based on whether the queue is associated with the first processor or with the second processor. Other embodiments are possible, and each of the embodiments can be used alone or together in combination.

Abstract: The present disclosure generally relates partial speed changes to improve in-order data transfer. Rather than determining an ECC decoder on a first available decoder basis, the ECC decoder may be based on the ECC decoder level. A memory device will have at least one FMU that has a syndrome weight (SW). The disclosure proposes assigning FMU's based on the SW rate. At the time the command is read, the data storage device determines which level of decoder will be assigned to the FMU. The determination will then be checked according to different system environment parameters to maintain performance or reduce power consumption. The arrangement allows a more flexible system design that can adapt according to the current system status.

Abstract: Apparatuses and methods of directly accessing a memory space of a storage device by a host are provided. In one embodiment, a method of driverless access of a non-volatile memory of a non-volatile memory device by a host includes initializing a PCIe memory space mapping a portion of the non-volatile memory of the non-volatile memory device to a host memory space. The non-volatile memory is mapped through a PCIe link between the host and the non-volatile memory device. Load/store commands are sent to the PCIe memory space for driverless access. The method further includes negotiating an alignment size of the minimum transaction packet size to complete the load/store commands.

Abstract: The present disclosure generally relates to improving adaptive tuning of different clock rates of a memory device. Rather than clock rates only being determined off of one parameter such as workload, the clock rates now will be determined using multiple parameters. The tuning may be based on system parameters to allow the system to withstand challenges that arise during the operation. The clock frequency table is maintained in the device controller. The table holds the clock frequency of each component. The disclosure proposes modifying the table according to different system environment parameters to maintain performance or reduce power consumption. Adaptive tuning allows a more flexible system design that can adapt according to the current system status. Adaptive tuning also reduces peak power consumption, improves performance, and better quality of service (QoS) compatibility characteristics.

Abstract: A data storage device includes a memory device and a controller coupled to the memory device. The controller is configured to compare a first copy of a boot partition to a second copy of the boot partition. The first copy of the boot partition and the second copy of the boot partition each comprises a same number of a plurality of boot chunks. The boot partition corresponds to data of a boot operation of a host device. The controller is further configured to mark one or more of the compared boot chunks that equals or exceeds a similarity threshold and update a reliability index based on the marking. Based on the marking and the reliability index, the controller may increase or decrease an amount of error correction needed for the boot data.