Seagate Technology Incorporated

Stephen H Perlmutter

Stacey Secatch

Steven Scott Williams

Benjamin J Scott

Kyumsung Lee

Andrew J Louder

Daniel John Benjamin

David W Claude

Ryan James Goss

Thomas V Spencer

Robert William Dixon

Matthew B Lovell

Jonathan M Henze

Matthew Stoering

A data storage system may have a plurality of memory cells located in different data storage devices that are arranged into a plurality of logical namespaces with each logical namespace configured to be sequentially written and entirely erased as a single unit. An asymmetry strategy may be proactively created with the asymmetry module in response to data access activity to the logical namespaces by the asymmetry module. A new mode, as prescribed by the asymmetry strategy, is entered for at least one logical namespace in response to an operational trigger being met. The new mode changes a timing of at least one queued data access request to at least one logical namespace.

Data storage system with workload-based asymmetry compensation

Apparatus and method for a die kill and recovery sequence for a non-volatile memory (NVM). Data are stored in the NVM as data sets in garbage collection units (GCUs) that span multiple semiconductor dies. A die failure management circuit is configured to detect a die failure event associated with a selected die, and to generate a recovery strategy to accommodate the detected die failure event by selecting recovery actions to be taken in a selected sequence to maintain a current level of data transfer performance with a client device. The selected recovery actions are carried out in the selected sequence to transfer at least a portion of the user data stored in the selected die to a new replacement die, after which the selected die is decommissioned from further use. The NVM may be a flash memory of a solid-state drive (SSD).

Semiconductor die failure recovery in a data storage device

Apparatus and method for managing a non-volatile memory (NVM) such as a flash memory in a solid-state drive (SSD). In some embodiments, the NVM is arranged as a plurality of semiconductor memory dies coupled to a controller circuit using a plurality of channels. The controller circuit divides the plurality of dies into a succession of garbage collection units (GCUs). Each GCU is independently erasable and allocatable for storage of user data. The GCUs are staggered so that each GCU is formed from a different subset of the dies in the NVM. In further embodiments, the dies are arranged into NVM sets in accordance with the NVMe (Non-Volatile Memory Express) specification with each NVM set addressable by a different user for storage of data in a separate set of staggered GCUs.

Staggered garbage collection unit (GCU) allocation across dies

A data storage device can be arranged with a semiconductor memory having a plurality of erasure blocks accessed by a controller to store data. An access count for each respective erasure block can be generated to allow a wear range for the semiconductor memory to be computed based on the respective access counts with the controller. A performance impact of the wear range is evaluated with the controller in order to intelligently alter a deterministic window of a first erasure block of the plurality of erasure blocks in response to the performance impact.

Data storage device with wear range optimization

Apparatus and method for storing data in a non-volatile memory (NVM), such as a flash memory in a solid-state drive (SSD). In some embodiments, a distributed storage space of the NVM is defined to extend across a plural number of regions of the NVM. A non-standard parity data set is provided having a plural number of data elements greater than or equal to the plural number of regions in the storage space. The data set is written by storing a first portion of the data elements and a first parity value to the plural number of regions and a remaining portion of the data elements and a second parity value to a subset of the plural number of regions. The regions can comprise semiconductor dies in a flash memory, and the distributed storage space can be a garbage collection unit formed using one erasure block from each flash die.

Providing additional parity for non-standard sized parity data sets

Apparatus and method for managing entropy in a cryptographic processing system, such as but not limited to a solid-state drive (SSD). In some embodiments, a processing device is operated to transfer data between a host device and a non-volatile memory (NVM). In response to the detection of a power down event associated with the processing device, entropy associated with the power down event is collected and stored in a memory. Upon a subsequent reinitialization of the processing device, the entropy is conditioned and used as an input to a cryptographic function to subsequently transfer data between the host device and the NVM. In some embodiments, the entropy is obtained from the state of a hardware timer that provides a monotonically increasing count for timing control. In other embodiments, the entropy is obtained from a RAID buffer used to store data to a die set of the NVM.

Collection of uncorrelated entropy during a power down sequence

Apparatus and method for managing metadata in a data storage device such as a solid-state drive (SSD). In some embodiments, a write stream is formed of user data blocks to be sequentially written to a non-volatile memory (NVM). An entry of a reverse directory footer is generated for each user data block in the write stream to describe a physical address in the NVM at which the corresponding user data block is to be stored. The entries are accumulated in a buffer memory until the total count of entries reaches a predetermined threshold and a complete footer data structure is formed. The complete footer data structure is thereafter inserted into the write stream for writing, with the data blocks, to the NVM. The complete footer data structure has an overall size that corresponds to an overall size of each of the user data blocks.

Reverse directory structure in a garbage collection unit (GCU)

A data storage system can receive a data write request to write data to a physical address of a non-volatile semiconductor memory prior to detecting an error while storing the write data to the physical address. The detected error is corrected with a monitor module connected to the non-volatile semiconductor memory and a counter associated with the physical address is incremented with the monitor module in response to the corrected error. The write data can be subsequently read to a host in response to a data read request.

Growing community of inventors

Louisville, CO, United States of America

Stephen H Perlmutter