Sandisk Technologies Inc.

Sandisk Enterprise IP LLC

Other

Johann George

Brian Walter O'Krafka

Manavalan Krishnan

Darpan Dinker

Aaron Keith Olbrich

Harihara Kadayam

Niranjan Patre Neelakanta

Frederic H Tudor

Evgeniy Firsov

Yan Li

Steven Theodore Sprouse

John R Mitchell

Ramesh Chander

Mark Vrieling

Evgeny H Firsov

Vladislav Bolkhovitin

Akshay Mathur

Patrick Chiu

Vishal Kanaujia

Ajit Kumar Tapse

Systems, methods and/or devices are used to perform memory-efficient mapping of block/object addresses. In one aspect, a method of managing a storage system having one or more storage devices includes a tiered data structure in which each node has a logical ID and entries in the nodes reference other nodes in the tiered data structure using the logical IDs. As a result, when a child node is updated and stored to a new location, but retains its logical ID, its parent node does not need to be updated, because the logical ID in the entry referencing the child node remains unchanged. Further, the storage system uses a secondary mapping table to translate the logical IDs to the corresponding physical locations of the corresponding nodes. Additionally, the secondary mapping table is cached in volatile memory, and as a result, the physical location of a required node is determined without accessing non-volatile memory.

Memory-efficient block/object address mapping

In response to receiving a request to perform a transaction with two or more memory operations on one or more tiered data structures, the memory controller: writes a start transaction record to the log stream including a transaction identifier corresponding to the transaction; and performs the two or more memory operations. For a first memory operation associated with a key, the memory controller: writes a new data object in a datastore; assigns, in a key-map, a location of the new data object to the key; maintains an old data object in the datastore whose location was previously assigned to the key; and writes an operation commit record to a log stream upon completion of the first memory operation. In accordance with a determination that the two or more memory operations are complete, the memory controller writes a transaction commit record to the log stream including the transaction identifier.

Methods and systems for efficient non-isolated transactions

Memory-efficient object address mapping in a tiered data structure

Systems, methods, and/or devices are used to store metadata in a storage system. In one aspect, a first user space module sends a logical memory request to a memory management module of a kernel space module. The logical memory request includes data and metadata. A second user space module obtains the metadata of the logical memory request. A storage engine of the second user space module determines, in accordance with the obtained metadata, a location in non-volatile memory for the data. A second user space module generates a physical memory request including an indication of the non-volatile memory for the data. The second user space module transmits the physical memory request to the kernel space memory management module.

User space data storage management

Content is transferred from a first non-volatile storage medium to a second non-volatile storage medium without reproduction. This is accomplished by reading first data stored in the first non-volatile storage medium from the first non-volatile storage medium to volatile memory, erasing the first data from the first non-volatile storage medium, and after erasing the first data from the first non-volatile storage medium, storing corresponding data in a second non-volatile storage medium. And after storing the corresponding data in the second non-volatile storage medium, the first data is erased from the volatile memory. The first data is not concurrently stored in the first non-volatile storage medium and the second non-volatile storage medium.

Digitally transferring content across media without reproduction

Content is lent, or temporarily transferred, without reproduction, by reading first data stored in a first non-volatile storage medium from the first non-volatile storage medium to a first volatile memory, and transmitting corresponding data from the first volatile memory to a second volatile memory. After the reading or transmitting, at least a portion of the first data is erased from the first non-volatile storage medium. The data and corresponding data are also erased from the first volatile memory and second volatile memory and are thus stored in volatile memory for a period of not more than transitory duration. The portion of the first data stored in the first non-volatile storage medium and the corresponding data stored in the second non-volatile storage medium are not concurrently stored. In some embodiments, the first data includes one or more respective keys, each key for decrypting a respective encrypted data chunk.

Digitally lending content without reproduction

Systems, methods and/or devices are used to coalesce metadata and data writes via write serialization with device-level address remapping. In one aspect, a method of managing a storage system having one or more storage devices includes a serialized write operation to the storage system, in which a serialization segment accumulates data objects and mapping information until the segment is full, at which time the serialization segment is written to the storage system in a single contiguous write. As a result, the number of I/O operations is decreased from a minimum of two (one to write data and one to write updated mapping information) to a single write operation. Further, if the serialization segment contains existing valid data prior to accumulating data objects and mapping information, the valid data is moved to the beginning of the serialization segment using either a remap or xcopy operation.

Coalescing metadata and data writes via write serialization with device-level address remapping

A storage device may be configured to determine a delay associated with execution of a read operation responsive to a read command for data stored at the storage device. The storage device may send a notification that indicates the delay, that includes data that indicates a duration of the delay, or both. In response to receiving the notification, an access device may be configured to generate a second read command for redundant associated with data stored at the storage device.

Read operation delay

A memory controller receives request(s) to perform a plurality of memory operations on a datastore. The memory controller batches, into a set of leaf node operations, memory operations of the plurality of memory operations that are associated with keys in a range of keys, where a leaf node in a tiered data structure stored in the datastore is assigned the range of keys. The memory controller determines whether a condition is satisfied, and, in accordance with a determination that the condition is satisfied: locates the leaf node in the datastore; stores a copy of the leaf node in a cache; modifies the cached copy of the leaf node according to the set of leaf node operations so as to obtain one or more modified leaf nodes; and writes the one or more modified leaf nodes to one or more locations in the datastore with one or more write operations.

Methods and systems for insert optimization of tiered data structures

The embodiments described herein are methods and systems to enhance the reliability and performance of a persistent datastore (e.g., non-volatile memory such as flash memory). The method includes generating a log entry associated with first write data. The method also includes generating a first record including the log entry, the first write data, and pointer to a second record different from the first record. The method further includes performing a single write operation that includes writing the first record to the persistent datastore.

Method and system for forward reference logging in a persistent datastore

A memory controller detects a request, with a key to a data object and a value, to perform an operation. The memory controller locates the data object by mapping the key to a location of the data object in a tiered data structure and identifies a sequence number of the data object. When the sequence number of the data object is greater than a sequence number of a snapshot of the tiered data structure, the memory controller replaces the data object with a modified data object, including the key and the value, and assigns the modified object a unique sequence number. When the sequence number of the data object is less than the sequence number of the snapshot, the memory controller inserts a new data object, including the key and the value, into the tiered data structure and assigns the new data object the unique sequence number.

Methods and systems for performing efficient snapshots in tiered data structures

A computer system detects a request to access a first data object stored in a tiered data structure, that includes internal nodes and leaf nodes, where data objects in the leaf nodes include unique key information and corresponding values, and the first data object is uniquely identified by a first key. In response to detecting the request to access the first data object, the computer system retrieves a leaf node that includes the first data object and identifies the first data object in the leaf node by combining unique key information of the first data object with a key prefix that is stored separately in the leaf node to generate a combined key and determining that the combined key matches the first key that uniquely identifies the first data object. After identifying the first data object, the computer system provides access to the first data object.

Compaction of information in tiered data structure

In the present disclosure, a persistent storage device includes both persistent storage, which includes a set of persistent storage blocks, and a storage controller. The persistent storage device stores and retrieves data in response to commands received from an external host device. The persistent storage device stores data, from a contiguous data block, to two or more sets of logical address blocks in persistent storage. The persistent storage device also retrieves data, corresponding to a contiguous data block, from two or more sets of logical address blocks in persistent. In both instances, the two or more sets of logical address blocks in persistent storage, in aggregate, are not contiguous.

Reads and writes between a contiguous data block and noncontiguous sets of logical address blocks in a persistent storage device

A persistent storage device includes both persistent storage, which includes a set of persistent storage blocks, and NVRAM, and in particular a set of NVRAM blocks. The persistent storage device also typically includes a storage controller. The persistent storage device, in addition to responding to commands to write data directly to and to read data directly from persistent storage blocks is also configured to write data to specified NVRAM blocks (e.g., specified by a host NVRAM write command) and to transfer data from a specified NVRAM block to a specified persistent storage block. As a result, multiple writes to a particular persistent storage block can be replaced with multiple writes to an NVRAM block and a subsequent single write to the particular persistent storage block. This reduces the number of writes to persistent storage and also reduces the number of corresponding block erase operations.

Persistent storage device with NVRAM for staging writes

Approaches for recovering a node of a distributed transactional system. When a recovering node is being brought on-line, a copy of at least a portion of a first data store maintained by an existing node is created without ceasing to process transactions against the first data store at the existing node. The recovering node creates a second data store based on the copy of the first data store. While the recovering node creates the second data store, the recovering node stores committed transaction data received from a plurality of nodes. The committed transaction data describes transactions committed against data stores maintained by the sender of the committed transaction data. The recovering node may thereafter replay, against the second data store, any transactions identified by the committed transaction data that are associated with a global transaction number that is more recent than is a reference commit number.

Growing community of inventors

Sunnyvale, CA, United States of America

Johann George