Abstract: A control apparatus that is connected to a host controller, a first storage unit, and a second storage unit, and that is configured to perform mirror control of the first storage unit and the second storage unit, allocates in order readout requests accepted from the host controller as a readout request corresponding to the first storage unit or a readout request corresponding to the second storage unit, reads out data from the first storage unit and the second storage unit based on at least one allocated readout request, and transmits the readout data to the host controller.

Abstract: Disclosed are various embodiments for improving the resiliency and performance for clustered memory. A computing device can mark a page of the memory as being reclaimed. The computing device can then set the page of the memory as read-only. Next, the computing device can submit a write request for the contents of the page to individual ones of a plurality of memory hosts. Subsequently, the computing device can receive individual confirmations of a successful write of the page from the individual ones of the plurality of memory hosts. Then, the computing device can mark the page as free in response to receipt of the individual confirmations of the successful write from the individual ones of the plurality of memory hosts.

Abstract: Methods, systems, and devices for generating memory array control signals are described. A timing component may be configured to generate signals for operating a memory array. The timing component may include first logic that indicates when input signals are different, second logic that indicates when at least one of the input signals has a particular state, and third logic that indicates when the input signals have the same state. The output of the second logic and third logic may be controllable by other input signals. An output of the timing component may be set by one of the input signals and reset by the other input signals using the first logic, second logic, and third logic.

Abstract: A method is used for uninterrupted data flushing in a storage system. An input/output (I/O) request is received. A pointer for a filling flushing work set (FWS) is atomically switching. A copy of the pointer for the filling FWS is saved, and the counter associated with a FWS referenced by the saved copy of the pointer is incremented. The saved copy of the pointer and the pointer for the filling FWS are compared, and upon determining that the saved copy of the pointer and the pointer for the filling are FWS equal, the I/O request is committed to a page descriptor ring and the counter associated with the FWS referenced by the saved copy of the pointer is decremented.

Abstract: A circuit interface includes one or more processors that generate opcodes, a plurality of interface control circuits, each including a respective processing element responsive to the opcodes generated by one or more processors. Each interface control circuit corresponds to a respective link of a plurality of links of a device-to-device interface (DDI), and each link of the plurality of links of the DDI is for transmitting or receiving signals from one or more sources or one or more destinations external to the circuit.

Abstract: An apparatus is described. The apparatus includes a memory controller having an interface to communicate with a memory. The memory controller comprising logic circuitry to specify one of multiple possible write values to the memory during a write operation with multiple bits of a command that is sent on a command address bus that emanates from the interface. The memory to write any one of the possible write values into its storage cells while the memory interface is in a power saving state wherein the specified one write value is not articulated by the memory controller on a data bus of the interface as part of the write operation.

Abstract: Systems, apparatuses, and methods related to image based media type selection are described. Memory systems can include multiple types of memory media. Data can be written in a type of memory media based on one or more settings applied to the data. A setting can be determined based on input received by a logic within the memory system. In an example, a method can include receiving, at logic within a memory system that comprising a plurality of memory media types, data from an image sensor coupled to the logic of the memory system, receiving input from a host, identifying one or more attributes of the data, analyzing the received input to determine an setting, generating the setting based on the analyzed input, and writing the data to a first memory media type of the plurality of memory media types based on the generated setting.

Abstract: Systems, methods and products for performing file retention operations in a system in which a content management system accesses multiple cloud-based data stores that are compliant with a common file transfer protocol, but use different vendor-specific protocols for file retention operations, where a retention framework is coupled between the content management system and the data stores. The content management system performs file access operations by accessing an SDK that generates requests for these operations which follow the common set of protocols. The content management system performs file retention operations by accessing a retention framework that uses the information received from the content management system to identify the targeted file, identify the data store in which the targeted file is stored, and generate a request for a retention operation that is configured according to the data-store-specific retention protocols which correspond to the identified data store.

Abstract: An information processing apparatus includes a first memory, a second memory, and a processor configured to: change a data write destination from the first memory to the second memory when a write speed of data on the first memory becomes less than a speed threshold value; and perform an optimization process on the first memory in a state where the data write destination is the second memory, the optimization process optimizing a state of the first memory in such a manner as to restore the write speed of data on the first memory.

Abstract: A method and a system for end-to-end modeling are provided. The method includes: determining a topological structure of a target-based end-to-end model, where the topological structure includes an input layer, an encoding layer, an code enhancement layer, a filtering layer, a decoding layer and an output layer; the code enhancement layer adds information of a target unit to a feature sequence outputted by the encoding layer, the filtering layer filters a feature sequence added with the information of the target unit; collecting multiple pieces of training data; and training parameters of the target-based end-to-end model by using the multiple pieces of the training data.

Abstract: The described embodiments provide a system for controlling an integrated circuit memory device by a memory controller. During operation, the system sends a memory-access request from the memory controller to the memory device using a first link. After sending the memory-access request, the memory controller sends to the memory device a command that specifies performing a timing-calibration operation for a second link. The system subsequently transfers data associated with the memory-access request using the second link, wherein the timing-calibration operation occurs between sending the memory-access request and transferring the data associated with the memory-access request.

Abstract: A multi-stream restore method for reducing restore time of file data to service an external read request for file data stored by a data deduplication system employs a per-file read-ahead cache that is populated by multiple parallel internal read-ahead streams. The number of parallel read-ahead streams is dynamically variable based upon processing conditions and a set of heuristically pre-determined criteria that characterize the file data including a data access pattern that is sequential and a processing load that is below a predetermined threshold. Processing conditions are periodically monitored and adjustments to the multi-stream data access are made to optimize performance.

Abstract: Resilient mediation between storage systems replicating a dataset, including: receiving, by a mediation service from one or more storage systems that synchronously replicate a dataset, a request to resolve which storage system continues to service a dataset after the request; and sending, from the mediation service to at least one of the storage systems, a positive mediation result, wherein: the storage systems that received the positive mediation result continue to process data storage requests directed to the dataset, and the storage systems that did not receive a positive mediation result from the mediation service do not continue to process data storage requests directed to the dataset.

Abstract: A data storage device includes a memory device and a controller coupled to the memory device. The controller is configured to receive read requests from a host device. When a read request is received corresponding to one or more pages from a first plane and one or more pages from a second plane, the controller is configured to determine a decode time for the pages of the first plane and the pages of the second plane. Based on the decode times for pages of the first plane and pages of the second page, pages of the first plane that have a similar decode time to pages of the second plane are read in parallel.

Abstract: Systems and methods are provided for managing the order of data written to a transaction log in a distributed storage system. In a system with multiple nodes, if sequencers are naively generated without taking into consideration inconsistencies among the different nodes generating the sequencers, then the sequencers may not increase for each data transaction. To alleviate this problem, the node committing the transaction to the transaction log may perform a consistent read and verifies that the sequencer advances. If the sequencer does not advance, the node can perform a context-dependent operation such as adjusting the sequencer, acknowledging the write without committing it to the transaction log, or rejecting the write altogether.

Abstract: A system, method, and computer program product for implementing data replication generation is provided. The method includes utilizing hardware and software resources within a hybrid cloud environment. A non-volatile memory host system and an associated target system are enabled for operational functionality and the non-volatile memory host system is connected to an I/O queueing component. In response, a plurality of queue structures is generated with respect to a host driver component and a connection between the non-volatile memory host system and the associated target system is detected. In response, a special purpose cache structure is generated and the plurality of queue structures and the special purpose cache structure are enabled such that remote data mirroring functionality is enabled.

Abstract: Methods and systems for managing data for startup purposes in a distributed system are disclosed. The distributed system may include any number of clients operably connected to a storage system via a connection fabric. By virtue of the connection topology, any of the clients may interact with the storage system to gain access to storage resources of the storage system. When a client is provided access to storage resources by the storage system, other clients may not be able to access the storage resources of the storage system.

Abstract: Shielding a storage device of a storage system from one or more storage performance enhancement procedures (SPEPs) is disclosed. A SPEP can be regarded as a procedure that manipulates data stored via a storage device in a manner that attempts to improve performance of the storage system. As such, SPEPs are generally distinct from basic reading and writing of data that is not associated with storage system performance optimization. A SPEP can typically be effectively run without storage device performance degradation, however some conditions, such as very full drives, etc., can result in substantial storage device performance degradation where one or more SPEP is performed. As such, shielding can restrict permission to perform one or more SPEPs at a storage device expected to, or actually experiencing, a threshold level of performance degradation. Shielding is distinct from conventional techniques marking a drive as failed, failing, etc., as is disclosed herein.

Abstract: A method includes receiving, at a direct memory access (DMA) controller of a memory device, a first command from a first cache controller coupled to the memory device to prefetch first data from the memory device and sending the prefetched first data, in response to receiving the first command, to a second cache controller coupled to the memory device. The method can further include receiving a second command from a second cache controller coupled to the memory device to prefetch second data from the memory device, and sending the prefetched second data, in response to receiving the second command, to a third cache controller coupled to the memory device.

Abstract: Scale-out of a controller and application migration consider the application status, laws and rules of a scale-out destination, and a resource usage status. If an operation status exceeds a threshold value, an administrative server for a computer system of a storage system calculates a first index for each execution target, including an application, a virtual machine, or a container operating in the computer system, based on information about cost, credibility, or performance when the execution target is migrated to, and caused to operate in, another computer system, and indexing information indicating laws or rules applied to the computer system using indexes. A second index indicating, as a dimensionless quantity, the selection degree of another computer system as a migration destination of the execution target is acquired. A migration destination of the execution target is decided from among other computer systems based on the first index and/or the second index.