METHOD FOR PROCESSING METADATA IN STORAGE DEVICE AND RELATED DEVICE

    公开(公告)号:EP4152163A1

    公开(公告)日:2023-03-22

    申请号:EP21823076.1

    申请日:2021-05-19

    摘要: Embodiments of this application disclose a metadata processing method in a storage device and a related device, and belong to the field of computer technologies. The method includes: a network interface card in the storage device receives an IO request, where the IO request includes a data read request or a data write request; the network interface card executes a metadata processing task corresponding to the IO request; and when determining that the metadata processing task fails to be executed, the network interface card requests a CPU in the storage device to execute the metadata processing task. It can be learned that embodiments of this application provide a method in which a network interface card and a CPU adaptively and dynamically execute a metadata processing task. In this way, some metadata processing tasks may be adaptively offloaded from the CPU to the network interface card, so that data processing pressure of the CPU can be reduced, and an excessive delay due to metadata processing by the network interface card can be further avoided. Accordingly, data read/write efficiency of the storage device is improved.

    DATA STORAGE SYSTEM, SMART NETWORK CARD, AND COMPUTING NODE

    公开(公告)号:EP4425320A1

    公开(公告)日:2024-09-04

    申请号:EP22897187.5

    申请日:2022-07-21

    IPC分类号: G06F3/06

    CPC分类号: G06F3/06 G06F12/06

    摘要: Embodiments of this application disclose a data storage system, an intelligent network interface card, and a compute node, and belong to the field of data storage. The data storage system includes a compute node and a plurality of storage nodes. The compute node stores metadata, and the metadata may indicate an address at which target data is stored in the plurality of storage nodes. In this way, the compute node may directly read the target data from a corresponding storage node based on the address indicated by the metadata stored in the compute node, does not need to perform cross-node data forwarding by using the plurality of storage nodes, and does not need to perform a plurality of times of metadata searching. This effectively reduces a data reading delay and improves a data reading speed.

    SERVICE SYSTEM, AND MEMORY MANAGEMENT METHOD AND APPARATUS

    公开(公告)号:EP4390680A1

    公开(公告)日:2024-06-26

    申请号:EP22866172.4

    申请日:2022-05-31

    IPC分类号: G06F9/455 G06F9/50 G06F12/02

    摘要: A service system, and a memory management method and apparatus are provided, and relate to the computer field. In this application, memory load of service nodes in a service system can be balanced, and losses in memory performance and costs can be avoided, so that use performance of memory resources is improved. The service system includes a plurality of service nodes. A memory of at least one of the plurality of service nodes is divided into a local resource and a global resource. The local resource is used to provide memory storage space for a local service node, the global resource of the at least one service node forms a memory pool, and the memory pool is used to provide memory storage space for the plurality of service nodes. When a specific condition is satisfied, at least a part of space in the local resource is transferred to the memory pool, or at least a part of space in the memory pool is transferred to the local resource.

    METADATA STORAGE METHOD AND DEVICE
    4.
    发明公开

    公开(公告)号:EP4120058A1

    公开(公告)日:2023-01-18

    申请号:EP21800042.0

    申请日:2021-04-30

    发明人: REN, Ren WANG, Chen

    IPC分类号: G06F3/06

    摘要: Embodiments of this application disclose a metadata storage method. The method may be applied to a storage system. The method in embodiments of this application includes: providing a first metadata storage unit and a second metadata storage unit, where a size of the first metadata storage unit is greater than a size of the second metadata storage unit; and when a size of target data reaches a specified threshold, storing metadata of the target data based on the first metadata storage unit; or when the size of the target data is less than the specified threshold, storing the metadata of the target data based on the second metadata storage unit. Embodiments of this application can reduce corresponding metadata during data storage and save storage resources.

    DATA DEDUPLICATION METHOD AND APPARATUS
    5.
    发明公开

    公开(公告)号:EP4016276A1

    公开(公告)日:2022-06-22

    申请号:EP20852903.2

    申请日:2020-07-27

    IPC分类号: G06F3/06 G06F16/174

    摘要: A deduplication method and apparatus are provided. In the method, a fingerprint record that includes a plurality of fingerprint record items is first obtained, and at least two first fingerprint record items that include a same fingerprint are then determined from the fingerprint record. For example, the at least two first fingerprint record items each include a first fingerprint, so that deduplication is performed on data corresponding to the first fingerprints in the at least two first fingerprint record items. The at least two first fingerprint record items are deleted, a stub of the first fingerprint is recorded in the fingerprint record, and the stub of the first fingerprint indicates that the first fingerprint is a duplicated fingerprint. Because a stub corresponding to the duplicated fingerprint is added to the fingerprint record, it can be directly determined, by using the stub, that a fingerprint included in the fingerprint record item is a duplicated fingerprint, so that the duplicated fingerprint can be quickly determined and deduplication is performed. This improves efficiency of a deduplication technology.

    DATA PROCESSING METHOD AND STORAGE DEVICE
    7.
    发明公开

    公开(公告)号:EP4030310A1

    公开(公告)日:2022-07-20

    申请号:EP20939891.6

    申请日:2020-12-14

    IPC分类号: G06F16/215 G06F3/06

    摘要: This application provides a data processing method and a storage device, and belongs to the field of storage technologies. In this application, the storage device performs deduplication and compression based on different grains, deduplicates data based on a large grain, and compresses the data based on a small grain. Therefore, a limitation that a deduplication grain and a compression grain need to be the same is removed. A deduplication ratio decrease caused by an excessively large grain and a compression ratio decrease caused by an excessively small grain are avoided to some extent, to improve an overall reduction ratio of deduplication and compression.