This disclosure describes techniques for improved multicast network telemetry implemented over multilayer switches in a PIM domain. The multilayer switches may be configured to collectively certify end-to-end flow provisioning, and to publish telemetry data certifying flow provisioning from a single notifier to an external controller host. Computational workload and network traffic for streaming data related to certifying path provisioning is kept to a minimum for each flow that needs to be certified, which also keeps compounding of network traffic for many different flows to a minimum. Moreover, since controller hosts are notified upon successful provisioning but not at other times, controller hosts can trust that the telemetric data is minimally latent, and may be relied upon to enact timely actions which produce desired outcomes.

Methods, systems, and devices for wireless communications are described. A method for wireless communication at a user equipment (UE) may include: receiving a multicast bandwidth part configuration for multicast communications between a base station and the UE, where the multicast bandwidth part configuration includes a maximum number of multiple-input, multiple-output layers to be used for the multicast communications; establishing a multicast link with the base station for multicast communications; determining to use a reduced number of antennas for multicast reception than the number of antennas used for unicast reception, or determining that the UE is able to decode the multicast communications prior to a remaining portion of the multicast communications; and receiving the multicast communications via the reduced number of antennas, or reducing the radio frequency power prior to completion of a number of repetitions or retransmissions of the multicast communications.

The present disclosure provides a network off-line model processing method, an artificial intelligence processing device and related products, where the related products include a combined processing device. The combined processing device includes the artificial intelligence processing device, a general-purpose interconnection interface, and other processing devices, where the artificial intelligence processing device interacts with the other processing devices to jointly complete computation designated by users. The embodiments of the present disclosure can accelerate the operation of the network off-line model.

A method may include bridging in, via a fabric, a multicast data packet from a source device to a first edge device of a plurality of edge devices and flooding the multicast data packet to the plurality of edge devices within a mutual subnetwork of the fabric. The method further includes bridging out the multicast data packet from a second edge device of the plurality of edge devices to a receiving device. The source device and the receiving device are located within the mutual subnetwork.

A gateway device and a configuration client for supporting selective forwarding of messages published to a group address or a virtual address in a wireless mesh network of communicatively coupled communication devices, such as a Bluetooth Mesh system. The configuration client maintains a mapping between unicast addresses of communication devices and group and virtual addresses in the network. The gateway device receives, from the configuration client, unicast addresses of those communication devices collectively identified by the group or virtual address in a received message. When the retrieved unicast addresses are all serviced by the gateway device, the message is transmitted by the gateway device on all interfaces corresponding to the communication devices addressed by the retrieved unicast addresses. When the retrieved unicast addresses are not all serviced by the gateway device, the message is transmitted by the gateway device on all except one of the interfaces.

Systems and methods for InfiniBand fabric optimizations to minimize SA access and startup failover times. A system can comprise one or more microprocessors, a first subnet, the first subnet comprising a plurality of switches, a plurality of host channel adapters, a plurality of hosts, and a subnet manager, the subnet manager running on one of the one or more switches and the plurality of host channel adapters. The subnet manager can be configured to determine that the plurality of hosts and the plurality of switches support a same set of capabilities. On such determination, the subnet manager can configure an SMA flag, the flag indicating that a condition can be set for each of the host channel adapter ports.

In one embodiment, a method includes receiving a broadcast, unknown-unicast, or multicast (BUM) frame from a connected device, where the BUM frame is associated with a broadcast domain, determining a segment within the broadcast domain associated with the device, adding to the BUM frame a segment identifier that uniquely identifies the segment within the broadcast domain, and causing the BUM frame to be delivered to one or more recipient network apparatuses in a network associated with the broadcast domain, where the segment identifier added to the BUM frame is configured to be used by the one or more recipient network apparatuses to selectively forward the BUM frame to connected devices that are associated with segment identifier.

In an example embodiment, a tool to provide calendar integration with video conferencing technology to allow for the rendering of a specialized graphical user interface during a video conference that displays not only an identification of which users are currently attending or otherwise connected to the video conference but also an identification of users that have been invited to the video conference but not yet attended.

A method for modifying packet data of a packet in a network device, where the method includes receiving, at an ingress pipeline of the network device, the packet, performing a lookup, in a packet translation ruleset, to compare the packet data to rule criteria of a rule in the packet translation ruleset, making a first determination that at least a portion of the packet data matches the rule criteria, and based on the first determination, adding a packet translation tag to the packet, where the packet translation tag includes a rule action, copying the packet translation tag and a portion of the packet to obtain a copied packet, modifying the copied packet as described in the rule action to obtain a modified copied packet, and forwarding the modified copied packet to an egress pipeline.

A method for offloading multicast replication from multiple tiers of edge nodes implemented by multiple host machines to a physical switch is provided. Each of the multiple host machines implements a provider edge node and a tenant edge node. One host machine among the multiple host machines receives a packet having an overlay multicast group identifier. The host machine maps the overlay multicast group identifier to an underlay multicast group identifier. The host machine encapsulates the packet with an encapsulation header that includes the underlay multicast group identifier to create an encapsulated packet. The host machine forwards the encapsulated packet to a physical switch of the network segment. The physical switch forwards copies of the encapsulated packet to tenant edge nodes at one or more ports that are determined to be interested in the underlay multicast group identifier.