Abstract: A method of error management includes, in response to a read request for first data from a first storage device of a plurality of storage devices under one or more common data protection schemes, receiving a read uncorrectable indication regarding the first data, obtaining uncorrected data and metadata of an LBA associated with the first data, and obtaining the same LBA from one or more other storage devices of the plurality. The method further includes comparing the uncorrected data with the data and metadata from the other storage devices, speculatively modifying the uncorrected data based, at least in part, on the other data to create a set of reconstructed first data codewords, and, in response to a determination that one of the reconstructed first data codewords has recovered the first data, issuing a write_raw command to rewrite the modified data and associated metadata to the first storage device.

Abstract: A method and a memory device are provided. A controller of the memory device retrieves a first command burst from a host memory. The first command burst includes a sequence of one or more commands from a first submission queue. The controller identifies that at least one command in the first command burst is at least one first fused command of a first set of fused commands. The first set of fused commands is to be processed atomically in the memory device, and a remainder of the first set of fused commands is on the first submission queue. The controller stores at least an identifier of the at least one first fused command in a set-aside buffer of the memory device.

Abstract: A method includes sending an enumeration of a resource unit of the computing device to a first computing system tenant and to a second computing system tenant. The enumeration is sent through a first protocol and indicating a managing protocol associated with managing the resource unit. The method further includes receiving a first request from the first computing system tenant to reserve the resource unit. The first request is received through the managing protocol. The method further includes receiving a second request from the second computing system tenant to reserve the resource unit. The second request is received through the managing protocol. The method further includes sending, to the second computing system tenant, an indication that the resource unit is reserved by the first computing system tenant. The indication is sent through the managing protocol.

Abstract: A method of error management includes, in response to a read request for first data from a first storage device of a plurality of storage devices under one or more common data protection schemes, receiving a read uncorrectable indication regarding the first data, obtaining uncorrected data and metadata of an LBA associated with the first data, and obtaining the same LBA from one or more other storage devices of the plurality. The method further includes comparing the uncorrected data with the data and metadata from the other storage devices, speculatively modifying the uncorrected data based, at least in part, on the other data to create a set of reconstructed first data codewords, and, in response to a determination that one of the reconstructed first data codewords has recovered the first data, issuing a write_raw command to rewrite the modified data and associated metadata to the first storage device.

Abstract: A computing system comprises a host system, a first storage device, a second storage device, a third storage device, a fabric interconnect device and a controller separate from the host system. The first, second, and third storage devices comprise a first, second, and third local memory buffer. The fabric interconnect device is configured to connect the first, second, and third storage devices over a fabric network to the host system. In response to receiving a write operation from the host system, a controller (e.g., on the first storage device or the fabric interconnect device) is configured to calculate error-correction data (e.g., parity data) by using data-protection operations (e.g., XOR operation(s)) directly on data stored on the first, second, and third local memory buffer, without having to rely on computing resources of the host system.

Abstract: A computing system comprises a host system, a first storage device, a second storage device, a third storage device, a fabric interconnect device and a controller separate from the host system. The first, second, and third storage devices comprise a first, second, and third local memory buffer. The fabric interconnect device is configured to connect the first, second, and third storage devices over a fabric network to the host system. In response to receiving a write operation from the host system, a controller (e.g., on the first storage device or the fabric interconnect device) is configured to calculate error-correction data (e.g., parity data) by using data-protection operations (e.g., XOR operation(s)) directly on data stored on the first, second, and third local memory buffer, without having to rely on computing resources of the host system.