A method performed by a central server node in a distributed machine learning environment is provided. The method includes: managing distributed machine learning for a plurality of local client nodes, such that a first set of the plurality of local client nodes are assigned to assist training of a first central model and a second set of the plurality of local client nodes are assigned to assist training of a second central model; obtaining information regarding network conditions for the plurality of local client nodes; clustering the plurality of local client nodes into one or more clusters based at least in part on the information regarding network conditions; re-assigning a local client node in the first set to the second set based on the clustering; and sending to the local client node a message including model weights for the second central model.

A method performed by a central server node in a distributed machine learning environment is provided. The method includes: managing distributed machine learning for a plurality of local client nodes, such that a first set of the plurality of local client nodes are assigned to assist training of a first central model and a second set of the plurality of local client nodes are assigned to assist training of a second central model; obtaining information regarding network conditions for the plurality of local client nodes; clustering the plurality of local client nodes into one or more clusters based at least in part on the information regarding network conditions; re-assigning a local client node in the first set to the second set based on the clustering; and sending to the local client node a message including model weights for the second central model.

Various implementations described herein refer to a device having a multi-layered logic structure with multiple layers including a first layer and a second layer arranged vertically in a stacked configuration. The device may have a first network that links nodes together in the first layer. The device may have a second network that links the nodes in the first layer together by way of the second layer so as to reduce latency related to data transfer between the nodes.

A client device is configured to communicate with an access point over a wireless network, exchanging data with the access point over a selected communication channel. The client device stores an identifier of the selected communication channel. After the wireless connection to the access point has ended, the client device initiates a process to reconnect to the access point over the selected communication channel using the stored identifier.

An information transmission method and apparatus are provided. The method comprises: a first node sends a first data packet to a second node, the first data packet carrying at least one of the following information: first identifier information, first sequence number information, and first timestamp information, wherein the first identifier information is used for instructing the second node returns a second data packet after receiving the first data packet; the first sequence number information is used for identifying the first data packet; and the first timestamp information is used for instructing the first node to send time information of the first data packet.

In a method for exchanging packets between first and second nodes of a packet-switched network, each packet comprises two fields settable to an idle value or measurement value. The first node transmits to the second node first packets having a filed set to measurement value. Upon reception of each first packet, the second node transmits back to the first node a second packet having a field set to measurement value. Upon reception of each second packet, the first node transmits to the second node a third packet having another field set to measurement value. A packet loss measurement is calculated as a difference between the number of first packets and the number of third packets.

In a method for exchanging packets between first and second nodes of a packet-switched network, each packet comprises two fields settable to an idle value or measurement value. The first node transmits to the second node first packets having a filed set to measurement value. Upon reception of each first packet, the second node transmits back to the first node a second packet having a field set to measurement value. Upon reception of each second packet, the first node transmits to the second node a third packet having another field set to measurement value. A packet loss measurement is calculated as a difference between the number of first packets and the number of third packets.

A method of managing a fill state of a buffer in an external device includes monitoring the latency of a network connection to an external device having a network buffer via a managing device. A state of fill of the network buffer is determined based on at least the monitored latency of the network connection, and the effective network speed is estimated based on the state of fill of the network buffer. One or more network traffic scheduling parameters are adjusted in response to the estimated effective network speed, such as a maximum currently usable network speed that is lower than a maximum possible speed of the network. The maximum currently usable network speed of the network connection is periodically increased if the monitored latency is in a normal state and the maximum currently usable network speed is lower than the maximum possible speed of the network.

A solution for route selection includes receiving, by a network repository, from a first network function (NF), a query related to a target NF; querying, by the network repository, a route selection node for a shortest path to the target NF; receiving, by the network repository, from the route selection node, an indication of the shortest path to the target NF; and based on at least receiving the indication of the shortest path to the target NF, transmitting, by the network repository, to the first NF, a route to the target NF. In some examples, the shortest path has at least one of: a minimum number of hops, a minimum latency, a minimum jitter, and a minimum weighted transport score. In some examples, the route selection node is co-located with the network repository, which may be a network repository function (NRF).

Non-limiting examples of the present disclosure describe proactive detection and notification of a potential loss of connectivity during an electronic collaborative communication. Subsequently, a state of the electronic collaborative communication is managed to improve, among other technical advantages, processing efficiency of associated computing devices and users involved in the electronic collaborative communication. A quality of a network feed for a participant in an electronic collaborative communication is identified and analyzed to generate a warning of potential loss of connectivity. A state of an electronic collaborative communication is managed relative to a continued quality evaluation of said network feed. For example, a network feed may be placed on hold and/or a communication suspended due to potential connectivity issues. A representation of a video feed may be updated and/or a communication resumed when a quality of that video feed is determined to satisfy a pre-selected quality threshold.