Method for federating a cluster from a plurality of computing nodes from different fault domains

    公开(公告)号:US12034600B2

    公开(公告)日:2024-07-09

    申请号:US16446288

    申请日:2019-06-19

    Abstract: The present disclosure describes a plurality of examples for federating a cluster from a plurality of interconnected computing nodes. The examples disclose a controller receiving network information and enclosure information associated with the computing nodes. The network information is indicative of a network topology between the computing nodes. The enclosure information is indicative of a configuration of an enclosure associated with a corresponding computing node. The controller identifies fault domains based on the network information and the enclosure information. Each fault domain of the fault domains includes one or more computing nodes impacted by at least one of a corresponding network fault event or a corresponding enclosure fault event, and the plurality of fault domains comprise first, second, and third fault domains. The controller selects computing nodes from the fault domains for federating the cluster. The selected computing nodes comprise a first computing node from the first fault domain and a second computing node from the second fault domain. The selecting of the computing nodes favors selecting the first computing node from the first fault domain with a smaller quantity of computing nodes over the third fault domain with a larger quantity of computing nodes. The controller allocates the selected computing nodes in federating the cluster to execute workloads.

    Assigning outlier-related classifications to traffic flows across multiple time windows

    公开(公告)号:US12027044B2

    公开(公告)日:2024-07-02

    申请号:US17515309

    申请日:2021-10-29

    CPC classification number: G08G1/0145 G08G1/0133 G08G1/0141

    Abstract: Systems and methods are provided for combining a multiple sub-time window sampling architecture with machine learning to detect outlier traffic flow behavior which may indicate malicious/problematic network activity. For example, a network device may obtain a sample of traffic flow data during a defined time window. The sample of traffic flow data may comprise information associated with a sampled subset of traffic flows transferred by a network device in the defined time window. The network device may partition the defined time window into two or more sub-time windows. In each sub-time window, using machine learning, the network device may assign an outlier-related classification to each sampled traffic flow based on the relative behavioral characteristics of all the sampled traffic flows. The network device may aggregate the outlier-related classifications for each sampled traffic flow across multiple sub-time windows, and process traffic flows based on the aggregated outlier-related classifications.

    Backing up data for a namespace assigned to a tenant

    公开(公告)号:US12026066B2

    公开(公告)日:2024-07-02

    申请号:US18328030

    申请日:2023-06-02

    Abstract: Examples described herein are generally directed to backing up and restoring of container clusters. According to an example, the conventional tight coupling between namespaces and tenants is eliminated by providing within a container platform a tenant abstraction for each tenant of the container platform that shares resources of a container cluster. The tenant abstraction for a given tenant includes information indicative of a subset of namespaces of the container cluster assigned to the given tenant. Responsive to receipt via a user interface of the container platform of a request to backup a particular tenant of the container platform: the container platform determines the subset of namespaces assigned to the particular tenant; and for each namespace, causes a namespace-level backup/recovery tool to backup object data for the namespace within a backup location, and backup data associated with persistent volume claims for each unit of cluster replication within the namespace.

    REDUCING INTERFERENCE BETWEEN CHANNELS OF WIRELESS WIDE AREA NETWORK AND WIRELESS LOCAL AREA NETWORK

    公开(公告)号:US20240214827A1

    公开(公告)日:2024-06-27

    申请号:US18145154

    申请日:2022-12-22

    CPC classification number: H04W16/14 H04W48/16

    Abstract: In implementations of the present disclosure, there is provided an approach for reducing interference between channels of a wireless wide area network (WWAN) and channels of a wireless local area network (WLAN). A method comprises determining a first working channel in a first frequency band of a WWAN, and selecting a first filter from a first set of filters corresponding to the first frequency band based on the first working channel. Then, a second filter from a second set of filters corresponding to a second frequency band in a WLAN and a second working channel in the second frequency band are selected. The determined first working channel and the selected second working channel are used to transmit data. Implementations of the present disclosure can help the terminal device to use channels of WWAN and WLAN simultaneously and reduce interference between channels of WWAN and WLAN.

    Journal groups for metadata housekeeping operation

    公开(公告)号:US12019620B2

    公开(公告)日:2024-06-25

    申请号:US17586131

    申请日:2022-01-27

    CPC classification number: G06F16/2379 G06F16/285

    Abstract: Example implementations relate to metadata operations in a storage system. An example includes detecting, by a storage controller of a deduplication storage system, a housekeeping operation to perform updates to data of a set of container indexes of stored deduplicated data; in response to a detection of the housekeeping operation, the storage controller identifying, among the set of container indexes, a plurality of container indexes associated with a particular journal group, the particular journal group comprising a plurality of journals to store updates to data of the plurality of container indexes, respectively; and maintaining, by the storage controller, the particular journal group loaded in memory until all of the updates of the housekeeping operation to the plurality of container indexes have been stored in the respective journals of the particular journal group.

Patent Agency Ranking