Abstract: A data index establishment method obtains M raw data segments and a first-layer mapping relationship through segment fitting on raw data in a raw data table, and further obtains N first key segments and a second-layer mapping relationship through segment fitting on a first key of each raw data segment. In a subsequent data query process, a multi-layer mapping relationship is input from a to-be-queried key sequentially in descending order, until a location of the to-be-queried key in the raw data table is located.

Abstract: Embodiments of the present invention disclose a data processing solution. In this solution, a hard disk receives a call instruction sent by a server, where the call instruction includes a function identifier of a to-be-called function and an operation address; and the hard disk calls the function in the hard disk based on the function identifier, and performs, based on the function, an operation on data corresponding to the operation address.

Abstract: A data storage method implemented by a storage node in a distributed storage system includes receiving first data of a first strip in a first stripe, wherein the first stripe comprises a first plurality of strips, receiving second data of a second strip in a second stripe, wherein the second strip comprises a second plurality of strips, wherein a first logical address of the first data is the same as a second logical address of the second data, and generating a record indicating that the first data reaches the storage node before the second data.

Abstract: Embodiments of the present invention disclose a data processing solution. In this solution, a hard disk receives a call instruction sent by a server, where the call instruction includes a function identifier of a to-be-called function and an operation address; and the hard disk calls the function in the hard disk based on the function identifier, and performs, based on the function, an operation on data corresponding to the operation address.

Abstract: A communication method includes monitoring, by a shared agent, shared memory, wherein the shared memory is used by a first application, wherein the first application runs on a virtual device, wherein the virtual device is located on a host, wherein the shared memory belongs to a part of memory of the host and does not belong to memory specified by the host for the virtual device, and wherein the shared agent is disposed on the host independent of the virtual device, determining, by the shared agent, whether data of the first application is written to the shared memory, reading, by the shared agent, the data from the shared memory and sending the data to a second application in response to the data of the first application is written to the shared memory, wherein the second application is a data sharing party specified by the first application.

Abstract: Embodiments of the present invention disclose a data processing solution. In this solution, a hard disk receives a call instruction sent by a server, where the call instruction includes a function identifier of a to-be-called function and an operation address; and the hard disk calls the function in the hard disk based on the function identifier, and performs, based on the function, an operation on data corresponding to the operation address.

Abstract: Embodiments of the present invention disclose a data processing solution. In this solution, a hard disk receives a call instruction sent by a server, where the call instruction includes a function identifier of a to-be-called function and an operation address; and the hard disk calls the function in the hard disk based on the function identifier, and performs, based on the function, an operation on data corresponding to the operation address.

Abstract: A distributed storage system including at least one storage node, and at least one object storage device (OSD) is deployed on the at least one storage node. At least one data processing device and at least one metadata processing device that has a mapping relationship with the at least one data processing device are deployed on the at least one OSD. A target data processing module of an input/output (IO) operation is configured to determine a target metadata processing device corresponding to the target data processing device and an identifier of an OSD in which the target metadata processing device is located, send an access operation to the target metadata processing device, and access data of the IO operation based on the IO operation. The target metadata processing device is configured to access metadata of the IO operation based on the access operation.

Abstract: A solution for processing a stripe in a storage device is provided. Where at least one stripe unit not requiring garbage collection from each stripe of at least two stripes in the storage device is determined, each of the at least two stripes comprises a stripe unit requiring garbage collection and a stripe unit not requiring garbage collection; parity data of data in the determined stripe units not requiring garbage collection is computed and stored into a first idle stripe unit, where the first idle stripe unit and the determined stripe units not requiring garbage collection are in a new stripe in the storage device.

Abstract: The solution discloses a data storage method in a distributed storage system: A client generates data of a stripe and concurrently sends data of strips in the stripe to storage nodes corresponding to the strips, so that data interaction between the storage nodes is reduced, and write concurrency is improved; and the storage node receives the data of the strips sent by the client and records a sequence of data of strips corresponding to a same logical address, so that write performance of a distributed block storage system is improved and data consistency is ensured.

Abstract: A distributed storage system comprises a data storage node Nx that receives a write success response returned by a hard disk that provides the storage address. The data storage node Nx returns a write success response to a client. The client receives the write success response returned by the data storage node Nx, and the client returns a write completion response to a host.

Abstract: A data index establishment method obtains M raw data segments and a first-layer mapping relationship through segment fitting on raw data in a raw data table, and further obtains N first key segments and a second-layer mapping relationship through segment fitting on a first key of each raw data segment. In a subsequent data query process, a multi-layer mapping relationship is input from a to-be-queried key sequentially in descending order, until a location of the to-be-queried key in the raw data table is located.

Abstract: A communication method includes monitoring, by a shared agent, shared memory, wherein the shared memory is used by a first application, wherein the first application runs on a virtual device, wherein the virtual device is located on a host, wherein the shared memory belongs to a part of memory of the host and does not belong to memory specified by the host for the virtual device, and wherein the shared agent is disposed on the host independent of the virtual device, determining, by the shared agent, whether data of the first application is written to the shared memory, reading, by the shared agent, the data from the shared memory and sending the data to a second application in response to the data of the first application is written to the shared memory, wherein the second application is a data sharing party specified by the first application.

Abstract: A distributed storage system including at least one storage node, and at least one object storage device (OSD) is deployed on the at least one storage node. At least one data processing device and at least one metadata processing device that has a mapping relationship with the at least one data processing device are deployed on the at least one OSD. A target data processing module of an input/output (IO) operation is configured to determine a target metadata processing device corresponding to the target data processing device and an identifier of an OSD in which the target metadata processing device is located, send an access operation to the target metadata processing device, and access data of the IO operation based on the IO operation. The target metadata processing device is configured to access metadata of the IO operation based on the access operation.

Abstract: A data collation method in a storage array including reading first data from a first logical address without decompression, assigning, by a storage controller, a second logical address to the first data, storing the first data to the second logical address, establishing, by the storage controller, a mapping relationship between an address of a storage array and the second logical address, where the first logical address is mapped to a first physical address of a storage device, a length of the first physical address is equal to a length of first data, a length of the first logical address is equal to a length of second data, and the first data is compressed data of the second data, and receiving, by the storage controller, the first data from the storage device.

Abstract: A storage system includes a host, a check node, and multiple data nodes. The host groups multiple data write requests into multiple data write request sets according to data node identifiers included in the data write requests. Each data write request set includes multiple data write requests having a same data node identifier. When a size of all data write requests in each of a specified quantity of data write request sets reaches a preset data amount, the host calculates check data of the specified quantity of data write request sets. Further, the host sends each data write request set to a data node indicated by an identifier, and sends the check data to the check node. Therefore, data of a same type can be stored in a same storage node and the data can be read from only one storage node, thereby improving data reading efficiency.

Abstract: A method for storing data in a distributed block storage system, where a client generates data of a stripe, and concurrently sends data of strips in the stripe to storage nodes corresponding to the strips in order to reduce data exchange between the storage nodes, and improve write concurrency, thereby improving write performance of the distributed block storage system.

Abstract: A flash memory apparatus for controlling storage of data comprises a plurality of blocks for storing data and a controller. Each block includes a plurality of pages. The controller receives first target data and identifies a first block to store the first target data. Then, the controller divides the first target data into two parts. A size of a first part is equal to the available capacity of the first block. And a size of a second part is equal to the size of the first target data minus the size of the first part. The controller further determines a second block that has stored data but is not full. At last, the controller writes the first part into the first block and write the second part into the second block.

Abstract: Embodiments of the present invention disclose a data processing solution. In this solution, a hard disk receives a call instruction sent by a server, where the call instruction includes a function identifier of a to-be-called function and an operation address; and the hard disk calls the function in the hard disk based on the function identifier, and performs, based on the function, an operation on data corresponding to the operation address.

Abstract: A storage system includes a host, a check node, and multiple data nodes. The host groups multiple data write requests into multiple data write request sets according to data node identifiers included in the data write requests. Each data write request set includes multiple data write requests having a same data node identifier. When a size of all data write requests in each of a specified quantity of data write request sets reaches a preset data amount, the host calculates check data of the specified quantity of data write request sets. Further, the host sends each data write request set to a data node indicated by an identifier, and sends the check data to the check node. Therefore, data of a same type can be stored in a same storage node and the data can be read from only one storage node, thereby improving data reading efficiency.