Abstract: A method may include receiving, at a storage device, a command using a storage protocol, wherein the storage device is configured to execute a user program, and executing, at the storage device, a device program based on the command. The command may be a first command, and the method may further include receiving, at the storage device, using the storage protocol, a second command, and sending, from the storage device, using the storage protocol, information about the device program based on the second command. The method may further include sending, from the storage device, using the storage protocol, a list of device programs supported by the storage device based on the second command. The method may further include providing, by the storage device, output data from the device program. The providing may include sending, from the storage device, using the storage protocol, a log message.

Abstract: A Key-Value (KV) storage device is disclosed. The KV storage device may include storage for objects, each object including data associated with a key. A host interface layer may receive requests to read data associated with a key from the storage, to write data associated with a key to the storage, and a boot request to get boot data from the storage. A boot request processor may process the boot request using the storage.

Abstract: A method for scheduling input/output (I/O) commands is described. The method includes receiving, by an I/O scheduler, an I/O command from an application; generating, by the I/O controller, an I/O resource requirement based on the I/O command; determining, by a traffic controller, that an amount of available resources satisfies a criteria based on the I/O resource requirement; and sending, by the traffic controller, the I/O command to a queue in response to the criteria being satisfied.

Abstract: Computing resources may be optimally allocated for a multipath neural network using a multipath neural network analyzer that includes an interface and a processing device. The interface receives a multipath neural network. The processing device generates the multipath neural network to include one or more layers of a critical path through the multipath neural network that are allocated a first allocation of computing resources that are available to execute the multipath neural network. The critical path limits throughput of the multipath neural network. The first allocation of computing resources reduces an execution time of the multipath neural network to be less than a baseline execution time of a second allocation of computing resources for the multipath neural network. The first allocation of computing resources for a first layer of the critical path is different than the second allocation of computing resources for the first layer of the critical path.

Abstract: A message queue storage device includes: a non-volatile flash memory unit including one or more flash memory dies including one or more pages grouped into one or more flash blocks; a volatile memory; a data port; and a storage controller configured to: receive, via the data port, a message write command including a message and a queue identifier; identify a queue from one or more queues based on the queue identifier; determine that the message is a persistent message; select a write physical location in one or more pages of the flash memory dies in which to store the message; and store the message associated with the queue at the write physical location in the one or more pages of the non-volatile flash memory unit.

Abstract: A system is disclosed. The system may include a processor and a memory coupled to the processor. A storage device may also be coupled to the processor. The storage device may include a first interface and a second interface. The storage device may be configured to extend the memory. A mode switch may select a selected interface of the first interface and the second interface for a command issued by the processor.

Abstract: A method for transferring data may include writing, from a producing device, data to a storage device through an interconnect, determining a consumer device for the data, prefetching the data from the storage device, and transferring, based on the determining, the data to the consumer device through the interconnect. The method may further comprise receiving, at a prefetcher for the storage device, an indication of a relationship between the producing device and the consumer device, and determining the consumer device based on the indication. The method may further comprise placing the data in a stream at the storage device based on the relationship between the producing device and the consumer device. The indication may be provided by an application associated with the consumer device. Receiving the indication may include receiving the indication through a coherent memory protocol for the interconnect.

Abstract: A method for data encryption may include receiving input data, finding a delimiter in the input data, generating, based on a position of the delimiter in the input data, a portion of data using a part of the input data, and encrypting the portion of data. The input data may include a record, the delimiter indicates a boundary of the record, and the portion of data may include the record. The position of the delimiter may be in the part of the input data. Generating the portion of data may include generating the portion of data based on a subset of the part of the input data. The part of the input data may be a first part of the input data, and the position of the delimiter may be in a second part of the input data.

Abstract: A method for data compression may include scanning input data, performing, based on the scanning, a compression operation to generate compressed data using the input data, finding, based on the scanning, a delimiter in the input data, and generating, based on a position of the delimiter in the input data, a portion of data using the compressed data. The input data may include a record, the delimiter indicates a boundary of the record, and the portion of data may include the record. The generating may include generating the portion of data based on a portion size. The portion size may be a default portion size. The portion size may be based on a default portion size and a length of a match in the input data.

Abstract: A method for computational storage may include storing, at a storage device, a first portion of data, wherein the first portion of data may include a first fragment of a record, and a second portion of data may include a second fragment of the record, and appending the second fragment of the record to the first portion of data. The method may further include performing, at the storage device, an operation on the first and second fragments of the record. The method may further include determining that the first portion of data may include a first fragment of a record, and a second portion of data may include a second fragment of the record, wherein appending the second fragment of the record to the first portion of data may include appending, based on the determining, the second fragment of the record to the first portion of data.

Abstract: A method for computational storage may include storing, at a storage device, two or more portions of data, wherein a first one of the two or more portions of data comprises a first fragment of a record and a second one of the two or more portions of data comprises a second fragment of the record, and performing, by the storage device, an operation on the first and second fragments of the record. The method may further include performing, by the storage node, a second operation on first and second fragments of a second record. The operation may include a data selection operation, and the method may further include sending a result of the data selection operation to a server. The method may further include sending a result of a first data selection operation to a server.

Abstract: A method for memory access may include receiving, at a device, a first memory access request for a parallel workload, receiving, at the device, a second memory access request for the parallel workload, processing, by a first logical device of the device, the first memory access request, and processing, by a second logical device of the device, the second memory access request. Processing the first memory access request and processing the second memory access request may include parallel processing the first and second memory access requests. The first logical device may include one or more first resources. The method may further include configuring the first logical device based on one or more first parameters of the parallel workload. The method may further include allocating one or more first resources to the first logical device based on at least one of the one or more first parameters of the parallel workload.

Abstract: A method for computational storage may include receiving, at a storage device, a modified version of a portion of data, generating, at the storage device, a restored portion of data from the modified version of the portion of data, and performing, at the storage device, an operation on the restored portion of data. The method may further include receiving, at the storage device, a request to perform the operation on the portion of data. The generating may include decompressing the modified version of the portion of data. The generating may include decrypting the modified version of the portion of data. The method may further include sending, from the storage device, a result of the operation on the restored portion of data. The operation may include a filtering operation. The operation may include a scanning operation. The method may further include dividing data to generate the portion of data.

Abstract: Disclosed are systems and methods by which a storage device may process and return I/O commands to a host in the order in which the host provided the commands, thereby reducing host overhead, including but not limited to the following: receiving a first I/O command and a second I/O command, the first I/O command and the second I/O command being assigned a sequence tag, issuing the first I/O command and the second I/O command to one or more storage channels based on their respective sequence tags, collecting a command completion notice of the first I/O command or the second I/O command when the first I/O command or the second I/O command has been respectively completed; and issuing a command completion notification to a host based on the sequence tag of the associated completed first I/O command or the second I/O command.

Abstract: A storage device is described. The storage device may store data in a storage memory, and may have a host interface to manage communications between the storage device and a host machine. The storage device may also include a translation layer to translate addresses between the host machine and the storage memory, and a storage interface to access data from the storage memory. An in-storage monitoring engine may determine characteristics of the storage device, such as latency, bandwidth, and retention.

Abstract: A method of processing data in a system having a host and a storage node may include performing a shuffle operation on data stored at the storage node, wherein the shuffle operation may include performing a shuffle write operation, and performing a shuffle read operation, wherein at least a portion of the shuffle operation is performed by an accelerator at the storage node. A method for partitioning data may include sampling, at a device, data from one or more partitions based on a number of samples, transferring the sampled data from the device to a host, determining, at the host, one or more splitters based on the sampled data, communicating the one or more splitters from the host to the device, and partitioning, at the device, data for the one or more partitions based on the one or more splitters.

Abstract: A computing system includes: a storage device, coupled to central processing unit, includes: an in-storage processing engine configured to receive and manage application data from an application executed in a host computer, an in-storage processing coordinator, in the in-storage processing engine, configured to perform in-storage processing with formatted data, based on the application data, includes performing integer math operations, floating point math operations, Boolean operations, reorganization of data bits or symbols, and combinations thereof on the application data, and a data preprocessor, in the in-storage processing coordinator, configured to align the formatted data from the application data to return an in-storage processing output to the application for continued execution.

Abstract: A method for page management in a memory system may include allocating a page of a mirror memory, copying a valid page from a block of device memory at a device to the page of the mirror memory, remapping the valid page from the block of device memory to the mirror memory, and modifying the block of device memory. The method may further include copying the valid page from the mirror memory to a free page at the device, and remapping the valid page from the mirror memory to the free page at the device. The remapping may be performed using a memory coherent interface. The method may further include deallocating a portion of the mirror memory associated with the valid page based on copying the valid page from the mirror memory.

Abstract: A method may include receiving, from a process, a memory allocation request for a memory system comprising a first channel having a first channel utilization and a second channel having a second channel utilization, selecting, based on the first channel utilization and the second channel utilization, the first channel, and allocating, to the process, a page of memory from the first channel. The selecting may include selecting the first channel based on a balanced random policy. The selecting may include generating a ticket based on a random number and a number of free pages, comparing the ticket to a number of free pages of the first channel, and selecting the first channel based on the comparing. The selecting may include selecting the first channel based on a least used channel policy.

Abstract: According to one general aspect, an apparatus may include a regeneration-code-aware (RCA) storage device configured to calculate at least one type of data regeneration code for data error correction. The RCA storage device may include a memory configured to store data in chunks which, in turn, comprise data blocks. The RCA storage device may include a processor configured to compute, when requested by an external host device, a data regeneration code based upon a selected number of data blocks. The RCA storage device may include an external interface configured to transmit the data regeneration code to the external host device.