Samsung Electronics Co., Ltd.

Alibaba Group Holding Limited

Stmicroelectronics Gmbh

Jianjian Huo

Changho Choi

Hingkwan Huen

Derrick Tseng

Shu Li

Praveen Krishnamoorthy

Vikas Sinha

Jingpei Yang

Gunneswara Rao Marripudi

Janki Bhimani

Bo Kang

Harry Rogers

Indira Joshi

Ajit Yagaty

An interface for enabling a computer device to utilize data property-based data placement inside a nonvolatile memory device comprises: executing a software component at an operating system level in the computer device that monitors update statistics of all data item modifications into the nonvolatile memory device, including one or more of update frequencies for each data item, accumulated update and delete frequencies specific to each file type, and an origin of the data item; storing the update statistics of each of the data items and each of the data item types in a database; and intercepting all operations, including create, write, and update, of performed by applications to all the data items, and automatically assigning a data property identifier to each of the data items based on current update statistics in the database, such that the data items and assigned data property identifiers are transmitted over a memory channel.

Heuristic interface for enabling a computer device to utilize data property-based data placement inside a nonvolatile memory device

A system and method for satisfying Quality of Service (QoS) attributes for a stream using a storage device with multi-stream capability is described. The storage device may include memory to store data. A host interface may receive requests, some of which may be associated with a stream. A host interface layer may schedule the requests in a manner that may satisfy the QoS attribute for the stream.

Multi-stream SSD QOS management

An interface for enabling a computer device to utilize data property-based data placement inside a nonvolatile memory device comprises: executing a software component at an operating system level in the computer device that monitors update statistics of all data item modifications into the nonvolatile memory device, including one or more of update frequencies for each data item, accumulated update and delete frequencies specific to each file type, and an origin of the data item; storing the update statistics of each of the data items and each of the data item types in a database; and intercepting all operations, including create, write, and update, of performed by applications to all the data items, and automatically assigning a data property identifier to each of the data items based on current update statistics in the database, such that the data items and assigned data property identifiers are transmitted over a memory channel to the non-volatile memory device.

Multi-stream SSD QoS management

A distributed storage system node is disclosed. The distributed storage system node may include at least one storage device, which may act as the primary replica for data subject to an Input/Output (I/O) request. A cost analyzer may calculate a local estimated time required to complete the I/O request at the primary replica, and a remote estimated time required to complete the I/O request at a secondary replica of the data. An I/O redirector may direct the I/O request to either the primary replica or the secondary replica based on the local estimated time required and the one remote estimated time required.

IO redirection methods with cost estimation

Example embodiments are described for data property-based data placement inside a nonvolatile memory device performed by a storage controller of the nonvolatile memory device. In one aspect, the embodiments include: executing a software component on the computer device that detects at least one of an executing application and a hardware device connecting to the computing device; responsive to detecting the at least one executing application and the hardware device, searching, by the software component, a workflow repository to find a predetermined workflow associated with the at least one executing application and the hardware device, wherein the predetermined workflow associates predefined data property identifiers to different types of data items written to the nonvolatile memory device by the executing application or the hardware device; comparing, by the software component, activities of the at least one executing application and the hardware device to the predetermined workflow; and using the predetermined workflow to automatically assign the data property identifiers to the data items used by the application or the hardware device, such that the data items and assigned data property identifiers are transmitted over a channel to the nonvolatile memory device for storage wherein the nonvolatile memory device reads the data property identifiers and identifies which blocks of the nonvolatile memory device to store the corresponding data items, such that the data items having the same data property identifiers are stored in a same block.

Data property-based data placement in a nonvolatile memory device

One embodiment of the present invention provides a system for facilitating a distributed storage system. The system receives, by a first client-serving machine, a first request to write data. The system writes the data to a first persistent cache associated with the first client-serving machine, wherein a persistent cache includes non-volatile memory. The system records, in an entry in a global data structure, a status for the data prior to completing a write operation for the data in a storage server, wherein the status indicates that the data has been stored in the first persistent cache but has not yet been stored in the storage server.

Method and system for distributed storage using client-side global persistent cache

One embodiment facilitates data placement in a storage device. During operation, the system receives incoming data from multiple streams, wherein the incoming data includes a plurality of data parts. The system formats the data parts by inserting a header before and a tail after each data part, wherein a formatted data part includes a respective data part, the header, and the tail, and wherein the header and the tail indicate an order and a destination zone for the respective data part within a stream. The system sequentially writes the formatted data parts to a pre-allocated journal zone of a non-volatile memory of the storage device. The system concatenates the data parts in the order indicated by their respective header and tail. The system sequentially writes the concatenated data parts to the non-volatile memory based on the destination zone indicated by their respective header and tail.

System and method for quality of service assurance of multi-stream scenarios in a hard disk drive

One embodiment described herein provides a system for managing metadata for a shingled magnetic recording (SMR) drive. During operation, the system stores a metadata file in a memory of a host of the SMR drive; subsequent to the SMR drive performing a write operation, the system generates a metadata update, and sends the generated metadata update to a buffer associated with the SMR drive and the memory of the host. In response to a compaction condition being met, the system launches, in the memory of the host, a metadata compaction operation to generate a compacted metadata file based on the previously stored metadata file and the metadata update, and sends the compacted metadata file to the SMR drive.

System and method for data organization in shingled magnetic recording drive

A multi-streaming memory system includes a memory, and a processor coupled to the memory, the processor executing a software component that is configured to identify multiple attributes that are each related to logical block addresses (LBAs), and that each correspond to each of a plurality of streams of data writes, evaluate an importance factor for each of the attributes for each of the streams, and clustering two or more of the LBAs by assigning a stream ID to each of the LBAs based on all of the importance factors for each of the LBAs and the assigned stream.

Smart I/O stream detection based on multiple attributes

One embodiment facilitates a high-density converged storage system. During operation, the system receives, by a volatile memory of a storage device via a memory bus, data to be stored in a non-volatile memory of the same storage device. The system writes, by a controller of the storage device, the data directly to the non-volatile memory via the volatile memory, e.g., without using a serial bus protocol.

Method and system for high-density converged storage via memory bus

A system and method for satisfying Quality of Service (QoS) attributes () for a stream () using a storage device () with multi-stream capability is described. The storage device () may include memory () to store data. A host interface () may receive requests (), some of which may be associated with a stream (). A host interface layer () may schedule the requests () in a manner that may satisfy the QoS attribute () for the stream ().

A storage device. The device includes both low-latency persistent memory and higher-latency nonvolatile memory. The persistent memory may be used for write caching or for journaling. A B-tree may be used to maintain an index of write requests temporarily stored in the persistent memory. Garbage collection may be performed in the nonvolatile memory while write requests are being stored in the persistent memory.

Growing community of inventors

San Jose, CA, United States of America

Jianjian Huo