Accelerator Loading Method, System, and Apparatus

    公开(公告)号:US20190266006A1

    公开(公告)日:2019-08-29

    申请号:US16407934

    申请日:2019-05-09

    Abstract: An accelerator loading apparatus obtains an acceleration requirement, where the acceleration requirement includes an acceleration function and acceleration performance of a to-be-created virtual machine, determines an image that meets the acceleration function and the acceleration performance, and determines a target host in which an available accelerator that can load the image is located, and then sends an image loading command to the target host. The image loading command includes a descriptor of the image, and is used to enable the target host to load the image for the available accelerator. In the method, a target host that can create the virtual machine may be determined based on the acceleration function and the acceleration performance of the to-be-created virtual machine, and an image used for acceleration is loaded to an available accelerator of the target host, to implement dynamic accelerator loading and deployment.

    Virtual machine live migration method, apparatus, and system

    公开(公告)号:US11604670B2

    公开(公告)日:2023-03-14

    申请号:US16807334

    申请日:2020-03-03

    Abstract: A virtual machine live migration method, apparatus, and system relating to the field of communications technologies are disclosed. The method includes: in a process of live migration of a target virtual machine, obtaining pre-recorded storage location information corresponding to target service configuration information, where the target service configuration information is used by a target virtual acceleration component corresponding to the target virtual machine to perform service processing; reading the target service configuration information in a storage location corresponding to the storage location information; and sending the target service configuration information to a destination server, so that the destination server stores the target service configuration information. In the present disclosure, a service processing failure can be avoided.

    Multi-Path Fault Detection
    14.
    发明申请

    公开(公告)号:US20200151075A1

    公开(公告)日:2020-05-14

    申请号:US16734912

    申请日:2020-01-06

    Abstract: A method for implementing fault detection includes: instructing, by a detection device, a detected device to configure a detected path and a return path, where the detected path is a path from a first physical port of the detected device to a second physical port of the detected device via a target unit of the detected device, the return path is a path from the second physical port to the detection device, and the target unit is a VNF or an accelerator; sending a detection packet to the detected device through the first physical port; and when receiving the detection packet transmitted through the detected path and the return path, determining that the detected path is not faulty. According to the method, it can be further determined that the path that passes through the VNF or the accelerator is not faulty.

    NETWORK DEVICE, CONTROLLER, QUEUE MANAGEMENT METHOD, AND TRAFFIC MANAGEMENT CHIP

    公开(公告)号:US20190068516A1

    公开(公告)日:2019-02-28

    申请号:US16165419

    申请日:2018-10-19

    CPC classification number: H04L47/60 H04L47/50 H04L47/6215

    Abstract: This application describes a network device, a controller, a queue management method, and a traffic management chip. The method may be applied to a traffic management chip that uses an HQoS technology, and can include receiving a queue management instruction sent by a controller, where the queue management instruction includes an identifier of a first scheduler and an identifier of a first queue, and the first scheduler is one of multiple first-level schedulers. The method may also include controlling, according to the queue management instruction, scheduling of the first queue by the first scheduler, where a queue scheduled by the first scheduler belongs to a queue resource pool of the TM chip, and the queue resource pool includes at least one to-be-allocated queue. In this application, decoupling between queue allocation and the first-level schedulers is implemented, flexibility of queue allocation is improved, and utilization of queue resources is improved.

    Method for measuring available bandwidth and communications apparatus

    公开(公告)号:US11463338B2

    公开(公告)日:2022-10-04

    申请号:US17482042

    申请日:2021-09-22

    Abstract: A method for measuring available bandwidth includes a transmit end that sends a first probe sequence to a receive end, and receives a first available bandwidth measurement result from the receive end, the transmit end determines a second probe sequence based on the first available bandwidth measurement result, where the second probe sequence includes a group of probe packets with increasing sending rates, a range of the sending rates of the second probe sequence is the same as a range of the sending rates of the first probe sequence, the transmit end sends the second probe sequence to the receive end, and receives the second available bandwidth measurement result from the receive end, and the transmit end obtains the available bandwidth based on the second available bandwidth measurement result.

    Multi-path fault detection
    19.
    发明授权

    公开(公告)号:US11106517B2

    公开(公告)日:2021-08-31

    申请号:US16734912

    申请日:2020-01-06

    Abstract: A method for implementing fault detection includes: instructing, by a detection device, a detected device to configure a detected path and a return path, where the detected path is a path from a first physical port of the detected device to a second physical port of the detected device via a target unit of the detected device, the return path is a path from the second physical port to the detection device, and the target unit is a VNF or an accelerator; sending a detection packet to the detected device through the first physical port; and when receiving the detection packet transmitted through the detected path and the return path, determining that the detected path is not faulty. According to the method, it can be further determined that the path that passes through the VNF or the accelerator is not faulty.

Patent Agency Ranking