Abstract: A system for improving radiation tolerance of memory senses an amount of radiation exposure and, based on the sensed amount of radiation exposure, determines whether to perform one or more techniques for mitigating the effects of the radiation exposure. As an example, the system may perform a data refresh operation by re-writing data that has been corrupted by radiation, or the system may adjust the reference voltage used to read memory cells. In another example, the system may perform a fault repair operation by re-programming cells that have erroneously transitioned from a program state to an erase state. The system may selectively perform different radiation-mitigation techniques in a tiered approach based on the sensed amount of radiation in order to limit the adverse effects of the more invasive techniques.

Abstract: A storage system is provided, including: a host including a memory, the memory including a submission queue and a completion queue and being configured to store update frequency information; and a memory system configured to be connectable with the host and including a nonvolatile memory and a controller, the controller configured to control the nonvolatile memory, to receive the update frequency information of the submission queue from the host, to read command information stored in the submission queue based on the received update frequency information, and to execute controlling for the nonvolatile memory based on the command information.

Abstract: According to one embodiment, a memory system manages a plurality of first weights that correspond to the plurality of queues, and a plurality of second weights that correspond to the plurality of queues. The memory system selects a queue of a largest or smallest second weight, of the plurality of queues, as a queue of a highest priority, and starts execution of a command stored in the selected queue. The memory system updates the second weight corresponding to the selected queue by subtracting the first weight corresponding to the selected queue from the second weight corresponding to the selected queue or by adding the first weight corresponding to the selected queue to the second weight corresponding to the selected queue.

Abstract: According to one embodiment, a controller manages a first set of blocks and a second set of blocks. The controller allocates a second block included in the second set of blocks to a first block included in the first set of blocks. In response to receiving one or more write command specifying the first block, the controller writes data associated with the one or more received write commands to the second block in units of a second minimum write size. When the first block is filled with data that has been written to the first block and unwritten region remains in the second block, the controller deallocates the second block from the first block, and allocates the deallocated second block to a write destination block other than the first block.

Abstract: A method, computer system, and a computer program product for block prediction are provided. A computer receives a first retrieval request for retrieving data from storage blocks. The computer performs a cosine similarity comparison of the first retrieval request compared to prior data retrievals. The computer selects a matching data retrieval of the prior data retrievals. The matching data retrieval has a closest match to the first retrieval request based on the cosine similarity comparison. The computer identifies another storage block from the matching data retrieval as a predicted block for the first retrieval request. The computer transmits a prefetch request to prefetch data from the predicted block.

Abstract: An application and a plurality of types of storage in a distributed storage system are communicated with. A write instruction that includes a key-value pair that in turn includes a key and value is received from the application. The key-value pair is stored in a selected one of the plurality of types of storage where the selected type of storage is selected based at least in part on a size or access frequency of the key-value pair. A link to the stored key-value pair is stored, including by: generating a key hash based at least in part on the key from the key-value pair and selecting one of a plurality of rows in an extensible primary table in an index based at least in part on the key hash. If it is determined there is sufficient space, the link to the stored key-value pair is stored, in the selected row. If it is determined there is insufficient space, the key-value pair is stored in an overflow row in a secondary table.

Abstract: A data access system has host computers having front-end controllers nFE_SAN connected via a bus or network interconnect to back-end storage controllers nBE_SAN, and physical disk drives connected via network interconnect to the nBE_SANs to provide a distributed, high performance, policy based or dynamically reconfigurable, centrally managed, data storage acceleration system. The hardware and software architectural solutions eliminate BE_SAN controller bottlenecks and improve performance and scalability. In an embodiment, the nBE_SAN (BE_SAN) firmware recognize controller overload conditions, informs Distributed Resource Manager (DRM), and, based on the DRM provided optimal topology information, delegates part of its workload to additional controllers. The nFE_SAN firmware and additional hardware using functionally independent and redundant CPUs and memory that mitigate single points of failure and accelerates write performance.

Abstract: A system enables entities to access a single platform in order to utilize electronic data storage for storing different types of information. One or more computers may operate an electronic data storage processing network that entities can access when updating information in electronic data storage. The electronic data storage processing network may operate a plurality of electronic data storage processing modules, which can include an aggregator module, a formatter module, an operator signer module, and a validator module. Based on the specific use case for which electronic data storage is utilized, recordable data that is to be added to the electronic data storage can be processed by the appropriate aggregating, formatting, signing, and validating functions provided by the electronic data storage processing modules.

Abstract: Techniques described herein relate to a method for managing data protection services for data clusters. The method includes identifying, by a mapping module of a data protection manager, a first discovery event associated with a first data cluster; in response to identifying the first discovery event: obtaining first data cluster topology information associated with the first data cluster from the first data cluster; obtaining data cluster data protection information from a data cluster data protection information repository; mapping first protection policies to the first data cluster using the first data cluster topology information and the data cluster data protection information; updating data cluster topology information protection policy mappings using the using the first data cluster topology information and the data cluster data protection information; and initiating performance of first data protection services for the first data cluster based on the first mapped protection policies.

Abstract: Methods and systems for electronic storage are provided. A storage system comprises a plurality of storage system front ends, a plurality of storage system back ends, and a plurality of solid state drive (SSD) agents. Each storage system front end resides on a server of a plurality of servers. Each server of the plurality of servers comprises one or more storage system back ends of the plurality of storage system back ends. Each storage system front end is able to receive I/O requests and relay information associated with the I/O requests to a relevant storage system back end. The relevant storage system back end communicates metadata associated with the I/O request to an SSD via an SSD agent.

Abstract: A QLC based data storage device leverages a host memory buffer (HMB) to achieve QLC direct write that increases performance of the QLC data storage device and reducing or eliminating disadvantages associated with a QLC folding approach. In one example, the QLC based data storage device includes a controller configured to receive a request to write data to a non-volatile memory, determine whether the request is a sequential write operation, determine whether a HMB of the data storage device is enabled, determine whether a HMB allocation is successful for a quad-level cell direct write, and responsive to determining that the request is not the sequential write operation, the HMB of the data storage device is enabled, and the HMB allocation is successful for the quad-level cell direct write, perform a direct write operation in a quad-level cell block of the non-volatile memory.

Abstract: A system establishes spools that have spool regions in first memory. The system communicates first spool insertions into the spools, and from the spools to the storage devices, to write into a first partition of second memory of the storage devices. The first spool insertions include data and are optimized towards a first spool insertion size. The system communicates second spool insertions into the spools, and from the spools to the storage devices, to write into a second partition of the second memory of the storage devices. The second spool insertions include metadata and are optimized towards a second spool insertion size.

Abstract: A method includes allocating, via a tier allocation component, a first portion of data to a first tier memory component and writing the first portion of data to the first tier memory component in response to a first tier free list having an available entry. The method further includes evicting a second portion of data from the first tier memory component in response to the first tier free list being empty when the first portion of data is allocated to the first tier memory component and writing the first portion of data to the first tier memory component in response to evicting the second portion of data.

Abstract: Disclosed herein are system, method, and computer program product embodiments for utilizing non-RAM memory to implement a cloud storage system. An embodiment operates by receiving a request from an on-premises computer system to securely access a cloud drive by receiving an object specific template for an object. Based on the object specific template, an object specific plugin is selected, wherein the object specific plugin is configured to provide a connection to a cloud-based repository to obtain real time data for the object. An instance of the object is generated and communicated to a cloud plugin, wherein the cloud plugin is configured to communicate to a specific cloud drive through an HTTP client and further upload the instance of the object to the specific cloud drive.

Abstract: A computer-implemented method includes receiving, from a user account, an operation request to perform an operation on data stored at a distributed computing environment. The operation request includes an on-premises token associated with the user account. The method also includes extracting, from the on-premises token, an access scope associated with permissions of the user account for accessing the data stored at the distributed computing environment. The method also includes translating the extracted access scope into query parameters compatible with the data stored at the distributed computing environment. The method also includes determining, using the query parameters, whether the operation request can access the data stored at the distributed computing environment. When the operation request can access the data stored at the distributed computing environment, the method includes executing the operation request.

Abstract: A memory expander includes a memory device that stores a plurality of task data. A controller controls the memory device. The controller receives metadata and a management request from an external central processing unit (CPU) through a compute express link (CXL) interface and operates in a management mode in response to the management request. In the management mode, the controller receives a read request and a first address from an accelerator through the CXL interface and transmits one of the plurality of task data to the accelerator based on the metadata in response to the read request.

Abstract: Systems, apparatus, and methods are disclosed comprising receiving temperature information corresponding to a write temperature of at least one of multiple pages of non-volatile memory cells of a group of non-volatile memory cells, determining a statistical measure of temperature information for the group non-volatile memory cells using the received temperature information, and storing the determined statistical measure of temperature information for the group of non-volatile memory cells. The stored determined statistical measure of temperature information can be used to optimize or improve one or more storage system operations.

Abstract: According to one embodiment, in a semiconductor device, a first chip is electrically connected to a terminal to which a signal from a host is input. The first chip is electrically connected to a second chip and to a third chip in parallel with the second chip. The first chip includes a first buffer memory and a second buffer memory. The first buffer memory corresponds to the second chip. The second buffer memory corresponds to the third chip. The second chip includes a third buffer memory. The third chip includes a fourth buffer memory. A capacity of the first buffer memory is equal to or larger than a capacity of the third buffer memory. A capacity of the second buffer memory is equal to or larger than a capacity of the fourth buffer memory.

Abstract: Disclosed herein is a device equipped with flash memory, which includes memory in which at least one program is recorded and a processor for executing the program. The memory includes flash memory including a data area and a backup area, and the program divides data into two or more segments depending on whether the data can be stored in a single page and stores the same in the data area. The first segment is stored in a page along with a segment number, indicating the sequential position of the divided data, a segment offset, indicating the number of pages between the pages in which the current segment and the next segment are stored, the size of a data file name, the size of the data, and the file name. At least one additional segment may be stored in another page along with the segment number and segment offset thereof.

Abstract: Provided is a storage device which communicates with a host device and configured to set a secure mode of a plurality of commands different in kind. An operating method of the storage device includes receiving a secure request indicating a protection of a first command and a protection of a second command of the plurality of commands, from the host device; setting a secure mode of the first and second commands, based on the secure request; receiving a first request indicating a request to execute the first command, from the host device; outputting a first response indicating failure of the first command to the host device, based on the first request; receiving a second request indicating a request to execute the second command, from the host device; and outputting a second response indicating failure of the second command to the host device, based on the second request.