FUNCTION-AS-A-SERVICE (FAAS) MODEL FOR SPECIALIZED PROCESSING UNITS

    公开(公告)号:US20210287323A1

    公开(公告)日:2021-09-16

    申请号:US16817390

    申请日:2020-03-12

    Abstract: A server that includes a graphics processing unit (GPU) may receive, from a first application that is remote from the server, a first request to reserve a first number of cores of the GPU for a first amount of time. The server may also receive, from a second application that is also remote from the server, a second request to reserve a second number of cores of the GPU for a second amount of time that at least partly overlaps the first amount of time. The server may determine that the first request is associated with a higher priority than the second request and, in response, may reserve the first number of cores for the first amount of time for the first application. The server may send, to the first application, an indication that the first number of cores have been reserved as requested by the first application.

    DYNAMIC ALLOCATION AND RE-ALLOCATION OF LEARNING MODEL COMPUTING RESOURCES

    公开(公告)号:US20210286650A1

    公开(公告)日:2021-09-16

    申请号:US16818955

    申请日:2020-03-13

    Abstract: This disclosure describes techniques for improving allocation of computing resources to computation of machine learning tasks, including on massive computing systems hosting machine learning models. A method includes a computing system, based on a computational metric trend and/or a predicted computational metric of a past task model, allocating a computing resource for computing of a machine learning task by a current task model prior to runtime of the current task model; computing the machine learning task by executing a copy of the current task model; quantifying a computational metric of the copy of the current task model; determining a computational metric trend based on the computational metric; deriving a predicted computational metric of the copy of the current task model based on the computational metric; and, based on the computational metric trend, changing allocation of a computing resource for computing of the machine learning task by the current task model.

    Dynamic policy exchange
    203.
    发明授权

    公开(公告)号:US11115873B1

    公开(公告)日:2021-09-07

    申请号:US16809408

    申请日:2020-03-04

    Abstract: Dynamic policy mapping is provided via mapping, by an Access Point (AP), a plurality of applications to a set of privilege groups for Quality of Service (QoS) levels in a network; transmitting the mapping of the privilege groups to a client device; receiving packets from the client device including QoS markers; and in response to determining that the QoS markers received from the client device do not match the privilege groups for the packets, performing a corrective action on the client device, wherein the corrective action includes one or more of: disassociating the client device from the network; and retransmitting the set of privilege groups to the client device. In some embodiments, the privilege groups are transmitted before the client device is associated with the AP, enabling the client device to select what AP to associate with based on the privilege groups.

    ATTRIBUTE-BASED ENCRYPTION FOR MICROSERVICES

    公开(公告)号:US20210264047A1

    公开(公告)日:2021-08-26

    申请号:US17236936

    申请日:2021-04-21

    Abstract: The disclosed technology provides solutions that enable scalable and secure data retrieval between microservices by using microservice attributes to encrypt container based data stores. A process of the technology can include steps for: instantiating a first microservice and a second microservice in a cloud environment, wherein the first microservice is associated with a first attribute label and the second microservice is associated with a second attribute label, generating a first key based on the first attribute label and a second key based on the second attribute label, associating a first data store with the first microservice, wherein the first data store is encrypted using the first key, and associating a second data store with the second microservice, wherein the second data store is encrypted using the second key. Systems and machine readable media are also provided.

    Systems and methods for detecting access point impersonators

    公开(公告)号:US11057773B2

    公开(公告)日:2021-07-06

    申请号:US16544651

    申请日:2019-08-19

    Abstract: The present disclosure is related to systems and methods for detecting and disabling operations of access point impersonators in a network. In one aspect, a method includes receiving, at a network controller of a network, beacon reports from a plurality of access points; determining, by the network controller, one or more target areas with at least one potential access point impersonator operating therein; sending, by the network controller, on-demand beacon requests to one or more of the plurality of access points with corresponding coverage areas overlapping with the one or more target areas; receiving, at the network controller, responses to the on-demand beacon requests from the one or more of the plurality of access points; and determining, by the network controller, at least one access point impersonator based on the responses.

    Automated power-aware green radio resource management (RRM)

    公开(公告)号:US11019565B2

    公开(公告)日:2021-05-25

    申请号:US16548727

    申请日:2019-08-22

    Abstract: The disclosed technology relates to a process for managing power consumption of Wi-Fi access points. The solution is composed of three parts. A first part creates power profiles for each of the switches having access points within a computer network that detail how power consumption for an access point can be modified based on changing one or more elements associated with the access point. The second part creates a time-based radio constraint profile that details power requirements for an access point based on client devices connected to the access point within a pre-defined area to ensure that each client device can connect and stay connected with the access point. The third part utilizes machine learning to generate a model that identifies optimal settings for the access point in order to achieve a specific level of power consumption for the access point.

Patent Agency Ranking