Huawei Technologies Co., Limited

Vmware, Inc.

Sheng Fang

Bin Wang

Qi Liu

Jin He

Xiangjun Song

Bing Niu

Xiangdong Che

Zhibo Hu

Ka Zhang

Jingrong Xie

Lili Wang

Haibo Wang

Chen Zhang

Shuying Liu

Guoqi Xu

Zuliang Wang

Wenyan Li

Yaqun Xiao

Ruyun Zhang

Ren Tan

Hanlin Li

Fang Gao

Yihong Li

Xiugang Wei

Kaiqiang Shen

Bingqing Guo

Huizhi Wen

Gang Weng

Mingpu Wang

Jianxi Deng

A first network device obtains a first route, where the first route includes a first route prefix. The first network device advertises a second route, where the second route includes a subnet prefix and a path identifier, and a range of the subnet prefix includes a range of the first route prefix.

Route advertisement method and related device

A communication method and apparatus are provided, so that packets of different services can be sent, based on a segment routing (SR) technology, on the same path specified by an SR policy. This addresses the significant limitation in the implementation of dual-stack reconstruction, and enhances communication efficiency. In the method, after receiving a first packet, a first network device forwards the first packet to a second network device based on a path specified by an SR policy. A destination endpoint specified by the SR policy is the second network device. An identifier of the destination endpoint includes a first IP address of the second network device and a second internet protocol (IP) address of the second network device.

Communication method and apparatus

A packet sending method and a related device are disclosed. The method includes: A network device obtains a first packet, where an internet protocol extension header of the first packet includes a first identifier, and the first identifier indicates a service requirement corresponding to the first packet. The network device determines, based on the first identifier and a second identifier, path indication information corresponding to the first packet, where a destination address of the first packet includes the second identifier, and the path indication information herein includes one or more of a network slice identifier and a segment identifier list. The network device updates the first packet based on the path indication information to obtain a second packet, and sends the second packet.

Packet sending method, device, and system

An example method of offloading a compute task to an accelerator by an application executing in a host cluster includes: sending, by an application executing in an initiator host, the compute task to a virtual accelerator in the initiator host, the virtual accelerator paired with a hardware accelerator in a target host of the host cluster; forwarding, by the virtual accelerator, the compute task to the hardware accelerator in the target host over a network; and receiving, at the application via the virtual accelerator, a result of the compute task from the hardware accelerator over the network.

Virtual accelerators in a virtualized computing system

A computing environment can include a host system that maintains a guest system, and a hardware component configured to implement artificial intelligence ('AI') methods of processing and analyzing date. The guest system can provide a virtual computing environment that receives a request to implement an AI application, and utilize a framework and a guest library to convert data from the AI application into an intermediate representation (“IR”). The host system can receive the IR with a virtual device (“VD”), and utilize an IR backend to translate the IR into hardware operations for the hardware component. Translated hardware operations can be provided to, and carried out by, the hardware component to provide an implementation of the AI application. Results of the hardware operations can be transmitted from the VD of the host system to a VD driver of the guest system, virtualizing the hardware component relative to the guest system.

Virtualizing hardware components that implement AI applications

A service processing method includes that a first network device determines a second network device on a forwarding path of a packet based on a service requirement; the first network device sends a first message to the second network device, to establish a first connection to the second network device; and the first network device sends a second message to the second network device using the first connection, where the second message requests the second network device to assist in determining information related to the forwarding path. when determining the second network device is needed to assist the first network device for information related to the forwarding path of the packet, the first network device actively establishes a connection to the second network device, and requests, using the established connection, the second network device to assist the first network device in determining the information related to the forwarding path.

Service processing method and network device

A binding segment identifier processing method and a device are provided. The method includes a network device receives a message sent by a controller, where the message includes type information and a segment identifier, and the type information indicates that the segment identifier is a binding segment identifier BSID. The network device performs a processing action on the BSID based on the type information. According to embodiments of this application, the network device can identify the BSID, to resolve a technical problem that an incorrect processing result is caused when a segment identifier list of the network device includes the BSID.

Binding segment identifier processing method and device

A network device serving as a head node on a first path may determine, based on obtained first constraint information, whether the first path meets a constraint condition of a target service, and further determine whether to switch a packet forwarding path from the first path to another path.

Path switching method, device, and system

An example method of virtualizing a hardware accelerator in a host cluster of a virtualized computing system includes: commanding, at an initiator host in the host cluster, a programmable expansion bus device to reconfigure as a virtual accelerator based on specifications of a hardware accelerator in a target host of the host cluster; executing, in the programmable expansion bus device, software to emulate the virtual accelerator as connected to an expansion bus of the initiator host; receiving, at the programmable expansion bus device, compute tasks from an application executing in the initiator host; and sending, to the target host, the compute tasks for processing by the hardware accelerator.

A routing information processing method and apparatus. The method includes a first network device determines a first segment identifier corresponding to a second network device and a first locator. The first network device determines a second segment identifier corresponding to a third network device and a second locator. The first network device sends routing information that includes the first segment identifier and the second segment identifier. A fourth network device advertises mirror routing information specific to the first network device, where the mirror routing information includes the second segment identifier or the second locator.

Routing information processing method and apparatus

A multicast data packet processing method performed by an intermediate node of a multicast tree includes: receiving a first Bit Index Explicit Replication (BIER) packet including a first label; obtaining, according to the first label, a second label corresponding to a multicast tree including the intermediate node; and obtaining a second BIER packet according to the second label and the first BIER packet, and sending the second BIER packet which includes the second label.

Multicast data packet processing method, and apparatus

A packet loss processing method and a network device are provided. The method includes: A first node obtains a first forwarding label of a first packet, where the first packet is a discarded packet. The first node determines, based on the first forwarding label, that the first node does not have a LSP corresponding to the first forwarding label. The first node sends a first message to a second node, where the first message includes the first forwarding label, and the first message is used to indicate that the first node does not have the LSP corresponding to the first forwarding label. The second node may be, for example, a peer node of the first node. The first node sends the message to the peer node, to indicate that the first node does not have the LSP corresponding to the forwarding label.

Packet loss processing method and network device

A computing environment can include a host system that maintains a guest system, and a hardware component configured to implement artificial intelligence ('AI') methods of processing and analyzing date. The guest system can provide a virtual computing environment that receives a request to implement an AI application, and utilize a framework and a guest library to convert data from the AI application into an intermediate representation ('IR'). The host system can receive the IR with a virtual device (&#x201c;VD&#x201d;), and utilize an IR backend to translate the IR into hardware operations for the hardware component. Translated hardware operations can be provided to, and carried out by, the hardware component to provide an implementation of the AI application. Results of the hardware operations can be transmitted from the VD of the host system to a VD driver of the guest system, virtualizing the hardware component relative to the guest system.

A multicast data packet processing method in a point-to-multipoint (P2MP)-based bit index explicit replication (BIER) multicast tree, which includes an intermediate node configured to use P2MP to forward a BIER packet, includes receiving a first multicast data packet with a BIER header having a first label. The method further includes obtaining a first forwarding entry based on the first label. The first forwarding entry includes the first label, a first identifier identifying the multicast tree, and a second identifier instructing to perform P2MP forwarding. The method further includes obtaining, based on the first identifier, a second forwarding entry including the first identifier and a second label. The method further includes obtaining a second multicast data packet based on the second identifier, the second forwarding entry, and the first multicast data packet, and sending the second multicast data packet. The second multicast data packet includes the second label.

Growing community of inventors

Beijing, China

Sheng Fang