A data processing method implemented by a controller includes receiving a processing request from a specified node that carries identifiers of a plurality of computing nodes, where the plurality of computing nodes are configured to execute a specified calculation task, determining a target switching device from switching devices that are configured to connect to the plurality of computing nodes, and separately sending, to the target switching device and the specified node, routing information that indicates data forwarding paths between the plurality of computing nodes and the target switching device. The target switching device is configured to combine, based on the routing information, data reported by the plurality of computing nodes, and then send combined data to each computing node. The specified node is configured to send the routing information to each computing node, and each computing node may report data to the target switching device based on the routing information.

A data processing method implemented by a controller includes receiving a processing request from a specified node that carries identifiers of a plurality of computing nodes, where the plurality of computing nodes are configured to execute a specified calculation task, determining a target switching device from switching devices that are configured to connect to the plurality of computing nodes, and separately sending, to the target switching device and the specified node, routing information that indicates data forwarding paths between the plurality of computing nodes and the target switching device. The target switching device is configured to combine, based on the routing information, data reported by the plurality of computing nodes, and then send combined data to each computing node. The specified node is configured to send the routing information to each computing node, and each computing node may report data to the target switching device based on the routing information.

Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a method may include simulating a degradation of at least one of a first wireless communication link with a network or a different second wireless communication link with the network based on detecting that a first set of one or more packets, which were received via the first wireless communication link with the network and were added to a packet buffer, and a second set of one or more packets, which were received via the different second wireless communication link with the network and were added to the packet buffer, satisfy a threshold occupancy of the packet buffer. Numerous other aspects are provided.

A method and apparatus for transmitting a service flow based on a flexible Ethernet, where a bandwidth resource corresponding to a bundling group (BG) of a flexible Ethernet is divided into M timeslots, service data of a service flow is encapsulated in N timeslots in the M timeslots, and the method includes: when a first PHY in the BG fails, determining, based on a preconfigured first timeslot configuration table (TCT), a target timeslot (TTS) in the N timeslots that is mapped to the first PHY; searching the M timeslots for an idle timeslot (ITS) based on the first TCT; adjusting the first TCT when a quantity of ITSs is greater than or equal to a quantity of TTSs, so that all the N timeslots are mapped to PHYs other than the first PHY; and transmitting the service flow by using the mapped PHYs of the bundling group.

A method and apparatus for transmitting a service flow based on a flexible Ethernet, where a bandwidth resource corresponding to a bundling group (BG) of a flexible Ethernet is divided into M timeslots, service data of a service flow is encapsulated in N timeslots in the M timeslots, and the method includes: when a first PHY in the BG fails, determining, based on a preconfigured first timeslot configuration table (TCT), a target timeslot (TTS) in the N timeslots that is mapped to the first PHY; searching the M timeslots for an idle timeslot (ITS) based on the first TCT; adjusting the first TCT when a quantity of ITSs is greater than or equal to a quantity of TTSs, so that all the N timeslots are mapped to PHYs other than the first PHY; and transmitting the service flow by using the mapped PHYs of the bundling group.

Methods, systems, and computer programs are presented for processing Ethernet packets at a Field Programmable Gate Array (FPGA). One programmable integrated circuit includes: an internal network on chip (iNOC) comprising rows and columns; clusters, coupled to the iNOC, comprising a network access point (NAP) and programmable logic; and an Ethernet controller coupled to the iNOC. When the controller operates in packet mode, each complete inbound Ethernet packet is sent from the controller to one of the NAPs via the iNOC, where two or more NAPs are configurable to receive the complete inbound Ethernet packets from the controller. The controller is configurable to operate in quad segment interface (QSI) mode where each complete inbound Ethernet packet is broken into segments, which are sent from the controller to different NAPs via the iNOC, where two or more NAPs are configurable to receive the complete inbound Ethernet packets from the controller.

Methods, systems, and computer programs are presented for processing Ethernet packets at a Field Programmable Gate Array (FPGA). One programmable integrated circuit includes: an internal network on chip (iNOC) comprising rows and columns; clusters, coupled to the iNOC, comprising a network access point (NAP) and programmable logic; and an Ethernet controller coupled to the iNOC. When the controller operates in packet mode, each complete inbound Ethernet packet is sent from the controller to one of the NAPs via the iNOC, where two or more NAPs are configurable to receive the complete inbound Ethernet packets from the controller. The controller is configurable to operate in quad segment interface (QSI) mode where each complete inbound Ethernet packet is broken into segments, which are sent from the controller to different NAPs via the iNOC, where two or more NAPs are configurable to receive the complete inbound Ethernet packets from the controller.

In embodiments of this application, the network node may receive a first data packet, where the first data packet carries first traffic feature data of a first data flow to which the first data packet belongs. The network node may determine a data forwarding policy based on the first traffic feature data. Because the data forwarding policy includes a policy used when the network node forwards a data packet included in the first data flow, the network node may forward the data packet based on the data forwarding policy. It may be learned that, in this solution, there is no need for a controller to determine the forwarding policy and the network node to execute the forwarding policy. Instead, the network node may learn of the traffic feature data of the transmitted data flow in real time, and determine and use the data forwarding policy in a timely manner.

In embodiments of this application, the network node may receive a first data packet, where the first data packet carries first traffic feature data of a first data flow to which the first data packet belongs. The network node may determine a data forwarding policy based on the first traffic feature data. Because the data forwarding policy includes a policy used when the network node forwards a data packet included in the first data flow, the network node may forward the data packet based on the data forwarding policy. It may be learned that, in this solution, there is no need for a controller to determine the forwarding policy and the network node to execute the forwarding policy. Instead, the network node may learn of the traffic feature data of the transmitted data flow in real time, and determine and use the data forwarding policy in a timely manner.

A system and method to transmit frames from a first node to a second node over a plurality of radio links comprising a classifier to classify said frames according to one of a plurality of flow and a sequence number within said one of said plurality of flow and adding said flow and sequence number in a header of said classified frame a splitter receiving said classified frames from said classifier and distributing said classified frames on one of said plurality of radio links for transmission to said second node, a joiner receiving said classified frames and reordering them using an indexed sequence queue corresponding to each of said plurality of flows, a timer for waiting for frames missing in the sequence in one of said indexed sequence queue, wherein when said timer expires, if said frame has not arrived it is deemed lost and a forwarder to extract frames from said sequence queue to forward.