An information processing device include: a memory; and a processor coupled to the memory and configured to: receive an access request directed to an access target and sets the access request in any one of a plurality of pending entries each of which includes latency information; issue a command that corresponds to the access request; control issuance of the command on a basis of the latency information of the any one of the pending entries; set a value that indicates latency for the access request in the latency information of the any one of the pending entries; and subtract a predetermined value from the latency information of the any one of the pending entries for each unit of time.

A method, computer program product, and computing system for receiving a flush request for a metadata page stored in a storage array of a multi-node storage system. The flush request may be queued on a flush request lock queue on at least one node of the multi-node storage system. One or more flush requests may be processed, via multiple nodes of the multi-node storage system, on the metadata page based upon, at least in part, the flush request lock queue.

A method, computer program product, and computing system for assigning a plurality of unique sequential identifiers to a plurality of tablets in a cache memory system. One or more metadata deltas associated with a metadata page stored in a storage array may be written to the plurality of tablets in the cache memory system. Each metadata delta stored in at least one tablet of the plurality of tablets may be written to the metadata page stored in the storage array, thus defining one or more destage tablets. A largest unique sequential identifier from the plurality of unique sequential identifiers assigned to the one or more destage tablets, may be written to the storage array, thus defining a current tablet identifier for the metadata page.

A storage system having high performance and high reliability includes a non-volatile storage device, a storage controller configured to control data to be read and written from and to the storage device using a storage function; and a volatile memory. In the reading and writing, the storage controller generates a log and stores the log in a log memory, writes the log stored in the memory to the storage device, and collects a capacity of the storage area of the memory storing the log written to the storage device. In collecting a free area of the memory, the storage controller executes a base image saving method of writing in the storage device in units of storage areas having a plurality of logs and collecting a free area, and a garbage collection method of writing in the storage device in units of logs and collecting a free area.

An apparatus is provided comprising storage elements to store data blocks, where each data block has capability metadata associated therewith identifying whether the data block specifies a capability, at least one capability type being a bounded pointer. Processing circuitry is then arranged to be responsive to a bulk capability metadata operation identifying a plurality of the storage elements, to perform an operation on the capability metadata associated with each data block stored in the plurality of storage elements. Via a single specified operation, this hence enables query and/or modification operations to be performed on multiple items of capability metadata, hence providing more efficient access to such capability metadata.

For a non-volatile memory that uses hard bit and a soft bit data in error correction operations, architectures are introduced for the compression of the soft bit data to reduce the amount of data transferred over the memory's input-output interface. For a memory device with multiple planes of memory cells, the internal global data bus is segmented and a data compression circuit associated with each segment. This allows soft bit data from a cache buffer of a plane using one segment to transfer data between the cache buffer and the associated compression circuit concurrently with transferring data from a cache buffer of another plane using another segment, either for compression or transfer to the input-output interface.

The present disclosure may include a processor that uses idle caches as a backing store for a boot code. The processor designates a boot core and an active cache from a plurality of cores and a plurality of caches. The processor configures remaining caches from the plurality of caches to act as a backing store memory. The processor modifies the active cache to convert cast outs to a system memory into lateral cast outs to the backing store memory. The processor copies a boot image to the backing store memory and executes the boot image by the boot core.

The present disclosure provides a processing device including: a coarse-grained pruning unit configured to perform coarse-grained pruning on a weight of a neural network to obtain a pruned weight, an operation unit configured to train the neural network according to the pruned weight. The coarse-grained pruning unit is specifically configured to select M weights from the weights of the neural network through a sliding window, and when the M weights meet a preset condition, all or part of the M weights may be set to 0. The processing device can reduce the memory access while reducing the amount of computation, thereby obtaining an acceleration ratio and reducing energy consumption.

A memory controller circuit is disclosed which is coupleable to a first memory circuit, such as DRAM, and includes: a first memory control circuit to read from or write to the first memory circuit; a second memory circuit, such as SRAM; a second memory control circuit adapted to read from the second memory circuit in response to a read request when the requested data is stored in the second memory circuit, and otherwise to transfer the read request to the first memory control circuit; predetermined atomic operations circuitry; and programmable atomic operations circuitry adapted to perform at least one programmable atomic operation. The second memory control circuit also transfers a received programmable atomic operation request to the programmable atomic operations circuitry and sets a hazard bit for a cache line of the second memory circuit.

A method and system for maintaining command queue order are disclosed. According to certain embodiments, commands are read from a host, storing command queue IDs in an array that will keep the queue IDs in order. After having the queue IDs stored in the array, the commands are processed in the data storage device (DSD). After processing, the commands are provided to a completion order adjustment module that will order the commands in queue ID order for sequential commands to be returned to the host. In certain embodiments, for a sequential command, other commands of the same sequence are searched for the array and ordered with the sequential command. If a particular command of the sequence is not found, the completion order adjustment module will wait to transfer the sequence until each command of the sequence is found. For commands not part of a sequence, these commands are transferred to the host.