Abstract: A data processing method and a storage apparatus are disclosed. The data processing method includes: receiving, by an NVMe storage device, an NVMe write command sent by a host, where the NVMe write command carries a key and a value pointer, the value pointer points to first storage space, and the first storage space is used to store a value; obtaining, by the NVMe storage device, the key from the NVMe write command and a value length, and allocating second storage space to the value according to the value length, where the second storage space is in the NVMe storage device; and obtaining, by the NVMe storage device, the value from the host, and storing the value in the second storage space.

Abstract: A data processing method and a storage apparatus are disclosed. The data processing method includes: receiving, by an NVMe storage device, an NVMe write command sent by a host, where the NVMe write command carries a key and a value pointer, the value pointer points to first storage space, and the first storage space is used to store a value; obtaining, by the NVMe storage device, the key from the NVMe write command and a value length, and allocating second storage space to the value according to the value length, where the second storage space is in the NVMe storage device; and obtaining, by the NVMe storage device, the value from the host, and storing the value in the second storage space.

Abstract: A cascading board, a solid state drive (SSD) shared remote access system and method relate to the communications field and used to perform shared remote access to an SSD. One end of the cascading board is coupled to a remote direct memory access (RDMA) network, and another end is coupled to an SSD. The cascading board includes an access processing apparatus and a processor coupled to the access processing apparatus. The processor is configured to initialize the access processing apparatus and the SSD. The access processing apparatus is configured to implement RDMA to the SSD when receiving an access command from the RDMA network.

Abstract: A data processing method and a storage apparatus are disclosed. The data processing method includes: receiving, by an NVMe storage device, an NVMe write command sent by a host, where the NVMe write command carries a key and a value pointer, the value pointer points to first storage space, and the first storage space is used to store a value; obtaining, by the NVMe storage device, the key from the NVMe write command and a value length, and allocating second storage space to the value according to the value length, where the second storage space is in the NVMe storage device; and obtaining, by the NVMe storage device, the value from the host, and storing the value in the second storage space.

Abstract: A storage system includes a computing node and a switching device coupled to each other. The computing node sends a first network packet to the switching device, the first network packet carries a resource identifier, a first offset, and an input/output (I/O) command. The switching device determines an object storage device (OSD) corresponding to an I/O operation. The switching device generates a second offset according to the first offset and a size of the OSD, obtains a first network address of a first storage node accommodating a first OSD on which the I/O operation is to be performed and an identifier of the first OSD, generates a second network packet, and sends the second network packet to the first storage node. In the foregoing manner, a computation amount of the computing node can be decreased, and running efficiency of the computing node is improved.

Abstract: A data processing method and a storage apparatus are disclosed. The data processing method includes: receiving, by a non-volatile memory express (NVMe) storage device, an NVMe write command sent by a host, where the NVMe write command carries a key and a value pointer, the value pointer points to first storage space, and the first storage space is used to store a value; obtaining, by the NVMe storage device, the key from the NVMe write command and a value length, and allocating second storage space to the value according to the value length, where the second storage space is in the NVMe storage device; and obtaining, by the NVMe storage device, the value from the host, and storing the value in the second storage space.

Abstract: Embodiments of the present disclosure provide a storage system, including a distribution apparatus and a storage device; the storage device includes at least two storage controllers and multiple storage units, where each storage unit is configured with any one of the at least two storage controllers as a home storage controller; and the distribution apparatus includes a front-end interface and at least two back-end interfaces, where the front-end interface is configured to connect to a host device, and each back-end interface is connected to each storage controller in a one-to-one correspondence manner.

Abstract: Embodiments of the present disclosure provide a storage system, including a distribution apparatus and a storage device; the storage device includes at least two storage controllers and multiple storage units, where each storage unit is configured with any one of the at least two storage controllers as a home storage controller; and the distribution apparatus includes a front-end interface and at least two back-end interfaces, where the front-end interface is configured to connect to a host device, and each back-end interface is connected to each storage controller in a one-to-one correspondence manner. According to the technical solutions provided in the present disclosure, the distribution apparatus parses an IO read/write instruction, so that the IO read/write instruction can be accurately sent to the home storage controller, which avoids forwarding the IO read/write instruction between the storage controllers, thereby improving IO processing efficiency of the storage system.

Abstract: A storage system includes a computing node and a switching device coupled to each other. The computing node sends a first network packet to the switching device, the first network packet carries a resource identifier, a first offset, and an input/output (I/O) command. The switching device determines an object storage device (OSD) corresponding to an I/O operation. The switching device generates a second offset according to the first offset and a size of the OSD, obtains a first network address of a first storage node accommodating a first OSD on which the I/O operation is to be performed and an identifier of the first OSD, generates a second network packet, and sends the second network packet to the first storage node. In the foregoing manner, a computation amount of the computing node can be decreased, and running efficiency of the computing node is improved.

Abstract: A cascading board, a solid state drive (SSD) shared remote access system and method relate to the communications field and used to perform shared remote access to an SSD. One end of the cascading board is coupled to a remote direct memory access (RDMA) network, and another end is coupled to an SSD. The cascading board includes an access processing apparatus and a processor coupled to the access processing apparatus. The processor is configured to initialize the access processing apparatus and the SSD. The access processing apparatus is configured to implement RDMA to the SSD when receiving an access command from the RDMA network.

Abstract: A direct memory access (DMA) transmission control method and apparatus, where the method includes selecting a target channel for the target DMA task according to a priority corresponding to the target DMA task when a DMA transmission request for transmitting data of a target DMA task is received, querying a task type and a priority of another DMA task that has occupied a channel and a task type of the target DMA task when the other DMA task exists on the DMA channel, comparing the task type and the priority of the other DMA task that has occupied the channel with the task type and the priority of the target DMA task, and controlling data transmission on the DMA channel according to a comparison result. Hence, the urgent DMA task can be preferentially processed.

Abstract: A data processing method and a storage apparatus are disclosed. The data processing method includes: receiving, by a non-volatile memory express (NVMe) storage device, an NVMe write command sent by a host, where the NVMe write command carries a key and a value pointer, the value pointer points to first storage space, and the first storage space is used to store a value; obtaining, by the NVMe storage device, the key from the NVMe write command and a value length, and allocating second storage space to the value according to the value length, where the second storage space is in the NVMe storage device; and obtaining, by the NVMe storage device, the value from the host, and storing the value in the second storage space.

Abstract: A direct memory access (DMA) transmission control method and apparatus, where the method includes selecting a target channel for the target DMA task according to a priority corresponding to the target DMA task when a DMA transmission request for transmitting data of a target DMA task is received, querying a task type and a priority of another DMA task that has occupied a channel and a task type of the target DMA task when the other DMA task exists on the DMA channel, comparing the task type and the priority of the other DMA task that has occupied the channel with the task type and the priority of the target DMA task, and controlling data transmission on the DMA channel according to a comparison result. Hence, the urgent DMA task can be preferentially processed.

Abstract: Embodiments of the present disclosure provide a storage system, including a distribution apparatus and a storage device; the storage device includes at least two storage controllers and multiple storage units, where each storage unit is configured with any one of the at least two storage controllers as a home storage controller; and the distribution apparatus includes a front-end interface and at least two back-end interfaces, where the front-end interface is configured to connect to a host device, and each back-end interface is connected to each storage controller in a one-to-one correspondence manner. According to the technical solutions provided in the present disclosure, the distribution apparatus parses an IO read/write instruction, so that the IO read/write instruction can be accurately sent to the home storage controller, which avoids forwarding the IO read/write instruction between the storage controllers, thereby improving IO processing efficiency of the storage system.

Abstract: Embodiments of the present invention provide a method, an apparatus, and a system for performing time synchronization on PCIE (PCI Express, peripheral component interconnect express) devices. The method mainly includes: a PCIE device receiving, through a hardware interface, a time synchronization signal sent from a clock source device; parsing, by the PCIE device, the time synchronization signal to obtain clock information carried in the time synchronization signal, and using the clock information as a clock of the PCIE device. The PCIE devices are supported to access a synchronous network, and the PCIE devices are supported to be used as a global clock source.

Abstract: Embodiments of the present invention provide a method, an apparatus, and a system for performing time synchronization on PCIE (PCI Express, peripheral component interconnect express) devices. The method mainly includes: a PCIE device receiving, through a hardware interface, a time synchronization signal sent from a clock source device; parsing, by the PCIE device, the time synchronization signal to obtain clock information carried in the time synchronization signal, and using the clock information as a clock of the PCIE device. The PCIE devices are supported to access a synchronous network, and the PCIE devices are supported to be used as a global clock source.