Inbound packets can be received by a network device that determines a receive pipeline latency metric based on a plurality of receive pipeline residency times of the inbound packets and determines a receive queue latency metric based on a plurality of receive queue residency times of the inbound packets. The receive queue latency metric and the receive pipeline latency metric can be reported to a data collector. The network appliance may also receive a plurality of outbound packets on a transmit queue, determine a transmit queue latency metric based on the transmit queue residency times of the outbound packets, and determine a transmit pipeline latency metric based on the transmit pipeline residency times of the outbound packets. The outbound packets may be transmitted toward their destination. The transmit queue latency metric and the transmit pipeline latency metric can be reported to the data collector.

A switch in a slice-based network can be used to enforce quality of service (“QoS”). Agents can run in the switches, such as in the core of each switch. The switches can sort ingress packets into slice-specific ingress queues in a slice-based pool. The slices can have different QoS prioritizations. A switch-wide policing algorithm can move the slice-specific packets to egress interfaces. Then, one or more user-defined egress policing algorithms can prioritize which packets are sent out into the network first based on slice classifications.

Techniques for providing request throttling using proportional, integral, and exponential smoothing algorithms are disclosed. A distributed computing system can include a throttler engine that receives a plurality of requests targeting a software component within the distributed computing system. The throttler engine can aggregate the requests into a queue based on a time window. The throttler engine can determine a received request rate and a request rate limit for the software component and then compute a throttled request rate. The throttled request rate can include correction terms derived from proportional and integral computations and a correction term obtained from an exponential smoothing algorithm. The throttler engine can then provide throttled requests from the queue to the software component.

A memory-efficient technique for performing weighted round-robin load balancing in a distributed computing system is described. In one example of the present disclosure, a system can determine an offset to apply to a list of node identifiers based on a counter value. The system can select a subset of node identifiers from the list of node identifiers based on the offset. The system can then select a node identifier from the subset of node identifiers based on the counter value and a length of the subset of node identifiers. The system can transmit data to a node that corresponds to the node identifier and increment the counter value. The system can repeat this process any number of times to distribute data among a group of nodes in the distributed computing system.

A stateful service processing method and apparatus are provided, related to the field of communication technologies. The method includes: preprocessing a received first packet to obtain a coalescing message of the first packet; coalescing the first packet into a first queue based on the coalescing message of the first packet, where the first queue is used to coalesce packets of a first connection to which the first packet belongs, and the first connection is a connection in which a stateful service is located; when a preset condition is met, processing, based on a context of the first connection, a plurality of packets coalesced in the first queue to obtain a second packet, where the context of the first connection is an updated context obtained after a previous second packet of the first connection is obtained; and transmitting the second packet to the host.

A network device, associated with peer network devices, may receive policy information for a protocol; and compute a first update message based on information regarding a route associated with the policy information. The network device may determine that an upper utilization threshold for one or more of peer queues, associated with the peer network devices, is not satisfied; and write the first update message to the peer queues based on determining that the upper utilization threshold is not satisfied. The network device may compute a second update message based on the information regarding the route; determine that the upper utilization threshold for one or more of the peer queues is satisfied; and pause writing the second update message to the peer queues based on the upper utilization threshold being satisfied. The network device may permit the peer network devices to obtain data from corresponding ones of the peer queues.

Technologies for adaptive network packet egress scheduling include a switch configured to configure an eligibility table for a plurality of ports of the switch, wherein the eligibility table includes a plurality of rounds. The switch is further configured to retrieve an eligible mask corresponding to a round of a plurality of rounds of the eligibility table presently being scheduled and determine a ready mask that indicates a ready status of each port. The switch is further configured to determine, for each port, whether the eligible status and the ready status indicate that port is both eligible and ready, and schedule, in response to a determination that at least one port has been determined to be both eligible and ready, each of the at least one port that has been determined to be both eligible and ready. Additional embodiments are described herein.

Various examples of the present disclosure relate to a transmitter apparatus, device, method, and computer program, to a receiver apparatus, device, method, and computer program, and to corresponding source and destination devices and communication devices. The transmitter apparatus comprises a plurality of ports for data to be transmitted to a destination device, with each port being associated with a transmission data rate. The transmitter apparatus comprises processing circuitry configured to obtain data to be transmitted to the destination device via the plurality of ports. The processing circuitry is configured to multiplex the data to be transmitted to the destination device according to a weighted round-robin scheme to generate a multiplexed data stream. The weights of the weighted round-robin scheme are based on the transmission data rate of the respective port the data is obtained over. The processing circuitry is configured to transmit the multiplexed data stream to the destination device.

A switching system having input ports and output ports and comprising an input queued switch with virtual channels. Typically only one of these can, at a given time, access a given output port from among the output ports. Typically the input queued switch includes arbiter apparatus which controls the input ports and output ports to ensure that at least one input port, from among the input ports, transmits at most one cell at a time, and/or that at least one output port, from among the output ports, which receives a cell, receives that cell over only 1 virtual channel (VC) from among the virtual channels. The arbiter apparatus may function as a dispatch unit in which typically, at least one output port, from among the output ports, receives at most one cell at a time.

Embodiments of this application disclose a method and an apparatus for queue scheduling, to reduce a network latency in a packet transmission process. The method includes: A first device obtains a first packet balance when scheduling a first queue, where the first packet balance indicates a volume of packets that can be dequeued from the first queue; and the first device schedules a second queue based on the first packet balance.