Method and system involving degradation of non-volatile memory based on write commands and drive-writes

#With the significant growth in the amount of data being stored in data storage systems and criticality of the data, reliability of the data storage systems is crucial. For the reliability of the data storage systems, customers typically rely on the warranties provided by the manufacturers of the data storage systems that specify certain performance metrics and a lifetime of the data storage system.

#However, these performance metrics and lifetime warranties cannot be completely relied upon if the amount of data being written to the data storage systems and the frequency of the writes exceeds the specifications upon which the performance metrics and lifetime of the data storage systems are warrantied. Exceeding these specifications increases the likelihood of the data storage systems failing, and worse, places the critical data stored on these data storage systems by customers at risk.

#At least one aspect is directed to a solid-state storage device. The solid-state storage device includes a controller, and a non-volatile memory (NVM) device. The controller receives one or more host-write commands. The controller calculates, based on the one or more host-write commands, a number of drive-writes per day. The controller calculates, based on the number of drive-writes per day, an aggregated amount of degradation to a non-volatile memory component. The controller generates, using a machine-learned model, a probability of failure value based on data associated with one or more input parameters to the machine-learned model and the aggregated amount of degradation to the non-volatile memory component. The controller generates, based on the probability of failure value, an alert from an individual drive.

#FIG. 1 is a block diagram illustrating components of a data storage system according to aspects of the subject technology.

#FIG. 2 is a flowchart illustrating a process of determining whether a solid-state storage drive will fail according to aspects of subject technology.

#FIG. 3 is a flowchart illustrating a process of determining whether a solid-state storage drive will fail according to aspects of subject technology.

#The detailed description set forth below is intended as a description of various configurations of the subject disclosure and is not intended to represent the only configurations in which the subject disclosure may be practiced. The appended drawings are incorporated herein and constitute a part of the detailed description. The detailed description includes specific details for the purpose of providing a thorough understanding of the subject disclosure. However, it will be apparent to those skilled in the art that the subject disclosure may be practiced without these specific details. In some instances, structures and components are shown in block diagram form in order to avoid obscuring the concepts of the subject disclosure. Like components are labeled with identical element numbers for ease of understanding.

#The subject technology is directed to the prediction of failure or other operational degradation (such as latency degradation) of a solid-state storage drive. As discussed in detail below, the subject technology uses a machine-learned model to generate a probability of failure of a solid-state storage drive and generate an alert if the probability of failure satisfies a threshold probability of failure. The inputs to the machine-learned model include, but are not limited to, various solid-state storage drive parameter data and a throughput acceleration coefficient based on an actual of number of drive-writes per day of the solid-state storage drive and a predetermined number of drive-writes per day that the solid-state storage drive is designed to withstand.

#FIG. 1 is a block diagram illustrating components of a data storage system 100 according to aspects of the subject technology. Data storage system 100 may be an enterprise solid-state storage drive (SSD), hyperscale SSD, data-center SSD, consumer SSD and the like. Data storage system 100 may be included in an enterprise storage environment. As depicted in FIG. 1, data storage system 100 includes controller 110, volatile memory 120, non-volatile memory 130, host interface 140, machine-learned module 180, and comparison module 190. Controller 110 is configured to process requests received from host system 150 via host interface 140 to access data in non-volatile memory 130. The data access requests received from host system 150 may include write requests to store host data in non-volatile memory 130, read requests to retrieve host data stored in non-volatile memory 130, and erase requests to erase host data stored in non-volatile memory 130. Other types of requests, such as status requests or device management requests, may be received from host system 150 and processed by controller 110.