Apple Inc.

Vadim Khmelnitsky

Daniel Jeffrey Post

Nir Jacob Wakrat

Matthew John Byom

Kenneth Louis Herman

Hugo Fiennes

Anthony Fai

Tahoma Madrone Toelkes

Arjun Kapoor

Barry J Corlett

Hsiao Thio

Alexander C Sanks

Eric Brandon Tamura

Alexander Paley

Muhammad N Ashraf

Yuhua Liu

Systems and methods for managing non-volatile memory devices are provided. Embodiments discussed herein provide rapid restart protection for journaling system. The rapid restart protection prevents the NVM from experiencing memory saturation when the NVM system is being forced to handle multiple successive restarts.

Rapid restart protection for a non-volatile memory system

Systems and methods are disclosed for configuring a non-volatile memory ('NVM'). In some embodiments, each block of the NVM can include a block table-of-contents ('TOC'), which can be encoded (e.g., run-length encoded) and dynamically-sized. Thus, as user data is being programmed to a block, the size of a block TOC can be concurrently recalculated and increased only if necessary. In some embodiments, the NVM interface can use a weave sequence stored in the context information and at least one weave sequence associated with each page of a block to determine whether to replay across the pages of the block after system boot-up.

Systems and methods for configuring non-volatile memory

Systems and methods are provided for unmapping unused logical addresses at mount-time of a file system. An electronic device, which includes a non-volatile memory ('NVM'), may implement a file system that, at mount-time of the NVM, identifies all of the logical addresses associated with the NVM that are unallocated. The file system may then pass this information on to a NVM manager, such as in one or more unmap requests. This can ensure that the NVM manager does not maintain data associated with a logical address that is no longer needed by the file system.

Mount-time unmapping of unused logical addresses in non-volatile memory systems

Systems and methods are disclosed for logical block address ('LBA) bitmap usage for a system having non-volatile memory ('NVM'). A bitmap can be stored in volatile memory of the system, where the bitmap can store the mapping statuses of one or more logical addresses. By using the bitmap, the system can determine the mapping status of a LBA without having to access the NVM. In addition, the system can update the mapping status of a LBA with minimal NVM accesses. By reducing the number of NVM accesses, the system can avoid triggering a garbage collection process, which can improve overall system performance.

LBA bitmap usage

Systems and methods are disclosed for providing a weave sequence counter ('WSC') for non-volatile memory ('NVM') systems. The WSC can identify the sequence in which each page of the NVM is programmed. The &#x201c;weave&#x201d; aspect can refer to the fact that multiple blocks can be open for programming at once, thus allowing the pages of these blocks to be programmed in a &#x201c;woven&#x201d; manner. Systems and methods are also disclosed for providing a host weave sequence counter (&#x201c;HWSC&#x201d;). Each time new data is initially programmed to the NVM, this data can be associated with a particular HWSC. The HWSC associated with the data may not change, even when the data is moved to a new page (e.g., for wear leveling purposes and the like). The WSC and HWSC may aid in, for example, performing rollback, building logical-to-physical mappings, determining static-versus-dynamic page statuses, and performing maintenance operations (e.g., wear leveling).

Weave sequence counter for non-volatile memory systems

Systems and methods are provided for storing data to or reading data from a non-volatile memory ('NVM'), such as flash memory, using a metadata redundancy scheme. In some embodiments, an electronic device, which includes an NVM, may also include a memory interface for controlling access to the NVM. The memory interface may receive requests to write user data to the NVM. The user data from each request may be associated with metadata, such as a logical address, flags, or other data. In response to a write request, the NVM interface may store the user data and its associated metadata in a first memory location (e.g., page), and may store a redundant copy of the metadata in a second memory location. This way, even if the first memory location becomes inaccessible, the memory interface can still recover the metadata from the backup copy stored in the second memory location.

Metadata redundancy schemes for non-volatile memories

Systems and methods are disclosed for handling unclean shutdowns for a system having non-volatile memory ('NVM'). In some embodiments, the system can leverage from information obtained from index pages in order to efficiently reconstruct logical-to-physical mappings after an unclean shutdown event. In other embodiments, the system can reconstruct logical-to-physical mappings by leveraging from context information stored in a NVM. In further embodiments, context information can be used in conjunction with index pages to reconstruct logical-to-physical mappings after an unclean shutdown.

Handling unclean shutdowns for a system having non-volatile memory

Systems and methods are disclosed for mount-time reconciliation of data availability. During system boot-up, a non-volatile memory ('NVM') driver can be enumerated, and an NVM driver mapping can be obtained. The NVM driver mapping can include the actual availability of LBAs in the NVM. A file system can then be mounted, and a file system allocation state can be generated. The file system allocation state can indicate the file system's view of the availability of LBAs. Subsequently, data availability reconciliation can be performed. That is, the file system allocation state and the NVM driver mapping can be overlaid and compared with one another in order to expose any discrepancies.

Mount-time reconciliation of data availability

Systems and methods are disclosed for limiting power consumption of a non-volatile memory (NVM) using a power limiting scheme that distributes a number of concurrent NVM operations over time. This provides a 'current consumption cap' that fixes an upper limit of current consumption for the NVM, thereby eliminating peak power events. In one embodiment, power consumption of a NVM can be limited by receiving data suitable for use as a factor in adjusting a current threshold from at least one of a plurality of system sources. The current threshold can be less than a peak current capable of being consumed by the NVM and can be adjusted based on the received data. A power limiting scheme can be used that limits the number of concurrent NVM operations performed so that a cumulative current consumption of the NVM does not exceed the adjusted current threshold.

Methods and systems for dynamically controlling operations in a non-volatile memory to limit power consumption

Systems and methods are provided for initiating wear leveling on block-aligned boundaries for non-volatile memories ('NVMs'), such as flash memory. In some embodiments, an electronic device including the NVM may suspend the programming of data upon reaching the end of a dynamic block. The electronic device may then perform wear leveling on a low-cycled block of the NVM. The electronic device may thus be configured to copy static data from the low-cycled block to another block of the NVM. After wear leveling has completed, the memory interface can program a second portion of the data to a new dynamic block of the NVM. This way, the electronic device can improve the efficiency of garbage collection. In addition, the electronic device can decrease the programming time for user generated writes, the wearing of the NVM, and overall power consumption.

Initiating wear leveling for a non-volatile memory

Systems and methods are disclosed for efficient buffering for a system having non-volatile memory ('NVM'). A tree can be stored in volatile memory that includes a logical-to-physical mapping between a logical space and physical addresses of the NVM. When the amount of memory available for the tree is below a pre-determined threshold, a system can attempt to reduce the number of data fragments in the NVM, and consequently flatten a portion of the tree. The NVM interface may select an optimal set of entries of the tree to combine. Any suitable approach can be used such as, for example, moving one or more sliding windows across the tree, expanding a sliding window when a condition has been satisfied, using a priority queue while scanning the tree, and/or maintaining a priority queue while the tree is being updated.

Faster tree flattening for a system having non-volatile memory

In a managed memory subsystem, information associated with the memory subsystem is copied from volatile memory in the memory subsystem to host system memory. The copying can be over a standard interface. Responsive to memory subsystem power up from a powered down state or power loss, the information is copied from the host system memory back to the volatile memory in the memory subsystem, where the information can be used by the memory subsystem to perform memory operations. Transferring information from host system memory to volatile memory in a memory subsystem is faster and more power efficient than transferring the same information from non-volatile memory to volatile memory in the memory subsystem.

Growing community of inventors

Foster City, CA, United States of America

Vadim Khmelnitsky