Samsung Electronics Co., Ltd.

Soo-Young Ji

Sang-Hwa Jin

Min-Ho Kim

Dongouk Moon

Heeseok Eun

Donghun Lee

Younggeon Yoo

Seunghan Lee

Wonseb Jeong

Hyunjoon Yoo

Hee Seok Eun

Kyungkeun Lee

A computational storage device that is configured to be electrically connected to a host, comprising: a power management integrated circuit that is configured to distribute power; a storage that is configured to store data using a first voltage provided from the power management integrated circuit; a computing module that is configured to perform an operation that is allocated by the host using a second voltage provided from the power management integrated circuit; and a performance monitoring module that is configured to monitor a first workload, a first amount of computation, and/or a first power state of the storage and/or is configured to monitor a second workload, a second amount of computation, and/or a second power state of the computing module and is configured to send a request for additional power to the host based on a monitoring result that is generated by the performance monitoring module.

Computational storage devices and power management methods thereof

A storage device includes; a memory, a management circuit configured to manage an offloading program table and a count table, and a computing circuit configured to perform a processing operation using the offloading program table, the count table, and the memory. The management circuit is further configured to, in response to a first offloading program and a first offloading request, selectively store the first offloading program in the offloading program table in accordance with a determination of whether an offloading program identical to the first offloading program is stored in the offloading program table, and update the count table storing a first count indicating a remaining number of processing operations using the first offloading program.

Storage device and method performing processing operation requested by host

A storage device includes a non-volatile memory including a plurality of namespaces including a plurality of logical blocks; and a storage controller configured to, for each of the plurality of namespaces, check a capacity that is a number of allocable logical blocks, among the plurality of logical blocks, and a size that is a sum of a number of currently allocated logical blocks, among the plurality of logical blocks, and the number of allocable logical blocks, detect a first namespace, among the plurality of namespaces, by using the capacity and the size, and provide at least a portion of the allocable logical blocks of a second namespace, among the plurality of namespaces, to the first namespace.

Storage device and method of operating the same

The present disclosure provides methods, apparatuses, and servers for cache-coherence. In some embodiments, an apparatus includes a plurality of compute express link (CXL) devices, and a switch. Each CXL device of the plurality of CXL devices includes a memory in which a portion of the memory is allocated as a cache buffer, to which different cache eviction policies are allocated. The different cache eviction policies are modified according to a cache hit ratio of the cache buffer. The switch is configured to connect the plurality of CXL devices to each other.

Apparatus and method for cache-coherence

A storage device includes: a memory device; a memory controller; and a cooling unit configured to guide a flow of a cooling material to the memory controller, wherein the cooling unit includes a housing, a guide member, and a pump, wherein the housing covers the memory controller and includes a first point and a second point, wherein the first point is disposed at a first side of the housing, wherein the second point is disposed at a second side of the housing that is below the first side of the housing, wherein the guide member is attached to the housing and guides the flow of the cooling material from the first point toward the second point, and wherein the pump is configured to adjust an amount of the cooling material flowing from the first point to the second point.

Storage device and cooling system of the storage device with spirally guide members

Disclosed are a storage system and an operation method therefor. The storage system includes: a host system; and a plurality of storage sets configured to interface with the host system. At least one of the plurality of storage sets includes: a first memory region; a second memory region; and a third memory region, and the at least one of the plurality of storage sets is configured to move data stored in the third memory region to a selected memory region among the first memory region and the second memory region based on a data access feature.

Storage system and operation method therefor

An electronic device includes a host device and a plurality of storage devices. The host device includes a processor and a baseboard management controller (BMC). Each of the plurality of storage devices includes a storage controller and a micro controller unit (MCU). The processor and the storage controller support in-band communication, and the BMC and the MCU support out-of-band communication. The BMC receives monitoring data from the MCU of each of the plurality of storage devices based on the out-of-band communication. The processor allocates a first workload among one or more workloads to a first storage device among the plurality of storage devices, based on a monitoring data set including the monitoring data. The first storage device executes the first workload based on the in-band communication.

Electronic device and operating method thereof

A storage device includes a nonvolatile memory device, and a storage controller configured to control the nonvolatile memory device. The storage controller includes: (i) a quality of experience (QoE) manager configured to schedule a request received from an external user equipment, based on storage device information and network information, and (ii) a software-defined networking (SDN) manager configured to set a network transfer path to the external user equipment, which is associated with video data corresponding to the request.

Storage devices, methods of operating storage devices, and streaming systems including storage devices

The present disclosure provides methods and apparatuses for data loss prevention of a storage device. In some embodiments, the data loss preventing method includes receiving, from a host system, a query plan corresponding to necessary data to be stored in a volatile memory. The data loss preventing method further includes generating, based on the query plan, a data priority list corresponding to the necessary data. The data loss preventing method further includes selecting, based on the data priority list, at least one portion of the volatile memory, when a main power supplied by the host system drops to or below a power level threshold. The data loss preventing method further includes moving the necessary data to the at least one portion of the volatile memory.

Storage device preventing loss of data in situation of lacking power and operating method thereof

A computational storage device includes a control module, and a nonvolatile memory connected to the control module. The nonvolatile memory is configured to store a plurality of graph data elements, which comprises a plurality of nodes and a plurality of edges that connect at least a portion of the plurality of nodes to each other, in a first memory area and a second memory area each having a plurality of blocks and having different read speeds. The control module is configured to store a first graph data element of the plurality of graph data elements having a relatively high degree of association with one node of the plurality of nodes in the first memory area, and store a second graph data element of the plurality of graph data elements having a relatively low degree of association with the one node of the plurality of nodes in the second memory area.

Computational storage device, storage system including the same and operation method therefor

An apparatus including: a plurality of compute express link (CXL) devices each including a memory and a processor for processing works stored in the memory; and a switch configured to connect the CXL devices to each other, wherein a first CXL device among the plurality of CXL devices selects at least one second CXL device from at least some CXL devices of the plurality of CXL devices to distribute works stored in a memory of the first CXL device based on a usable capacity of a memory of the at least some CXL devices.

Apparatus and method for distributing work to a plurality of compute express link devices

A storage device includes a controller configured to receive a pre-processing instruction command from an external device, a non-volatile memory configured to store an original graph data, and a buffer memory connected to the controller and the non-volatile memory, wherein the controller is configured to load the original graph data from the non-volatile memory, generate pre-processing graph data by classifying the original graph data depending on vector similarity in response to the pre-processing instruction command, generate metadata on the basis of the pre-processing graph data, and provide the pre-processing graph data and the metadata to the non-volatile memory, the non-volatile memory is configured to store the pre-processing graph data and the metadata in a data block, and the buffer memory is configured to buffer the original graph data, the pre-processing graph data, and the metadata.

Storage device for graph data

A storage device is provided. A storage device includes a non-volatile memory including a plurality of memory segments, and a storage controller connected to the non-volatile memory through a plurality of channels, each of the plurality of channels connected to a respective one of the plurality of memory segments such that each of the plurality of channels has a respective associated memory segment, wherein the storage controller is configured to generate parity according to speed information received from a host with respect to data to be written to the non-volatile memory and store the parity in at least one of the memory segments.

Storage device and operating method thereof

A matching key search method of a server system including a storage device, the storage device including a nonvolatile memory device and a buffer memory. The matching key search method includes receiving a search command for a keyword from a host, reading data from a memory area of the memory device corresponding to an address included in the search command and storing the read data in the buffer memory, determining whether the read data matches the keyword, transmitting a result of the determining to the host without transmitting the read data to the host, and generating a matching key table with reference to the result of the determining and storing the generated matching key table in the buffer memory.

Growing community of inventors

Suwon-si, South Korea

Soo-Young Ji