A method for controlling congestion in a datacenter network or server is described. The server includes a processor configured to host a plurality of virtual machines and an ingress engine configured to maintain a plurality of per-virtual machine queues configured to store received packets. The processor is also configured to execute a CPU-fair fair queuing process to control the processing of the packets by the processor. The processor is also configured to selectively trigger temporary packet per second packet transmission limits on top of a substantially continuously enforced bit per second transmission limit upon detection of a per virtual machine queue overload.

Systems and methods for port congestion resiliency in a Link Aggregation Group (LAG) including a multi-card LAG and/or a multi-switch LAG. A method includes receiving a packet for egress over the LAG; responsive to determining no congestion over internal ports not part of the LAG, hashing with all member ports in the LAG in a distribution state; and, responsive to determining congestion over the internal ports, hashing with only member ports on a local card in which the packet was received, wherein the hashing determines which member port the packet egresses from in the LAG. The multi-card LAG includes multiple cards where packets ingress and egress from, and the cards communicate via a backplane port which is not part of the LAG. The multi-switch LAG includes multiple chassis where packets ingress and egress from, and the chassis communicate via an inter-switch connectivity port which is not part of the LAG.

A method for interrupting a packet storm in a server is implemented by a baseboard management controller (BMC) included in the server and includes the steps of: assigning a setting value included in firmware of the BMC to a first value so as to enable receipt of specific packets from a network, the specific packets being transmitted using a specific routing scheme; determining whether a packet storm has occurred according to a number of the specific packets that are received; and assigning the setting value to a second value so as to disable receipt of the specific packets when it is determined that the packet storm has occurred.

An MC-LAG system may operate to monitor load conditions existing in two network switches, and to compute a load index value based on detected load conditions. If a computed load index value for a first switch is determined to exceed a predetermined threshold, an overloaded switch may predictively cause traffic to be routed to a second switch prior to rebooting of the first switch. Load index values may be computed based upon factors including excessive inter-switch link (“ISL”) flapping, excessive MAC flush or MAC move operations in a switch, excessive processing resource utilization in a switch.

A method includes a compute node transmitting data to a port of a first switch at a first data transfer rate, monitoring the temperature of the port, and a management node providing an instruction to the compute node in response to the port temperature exceeding a temperature limit, wherein the instruction instructs the compute node to reduce the first data transfer rate to the port. The method further includes the compute node reducing the data transfer rate to the port in response to receiving the instruction. The method is applicable to multiple compute nodes transmitting data to multiple ports of a first switch. The data transfer rate may be reduced by throttling the compute node, renegotiating a link speed between the compute node and the port, or redirecting data to another switch. The methods facilitate thermal control of a switch without its own thermal throttling capability.

In an apparatus of a communication network it is determined whether or not overload protection processing has been requested by a server, the overload protection processing being performed by dropping messages. In case it is determined that the overload protection processing has not been requested, a message to be sent from a client to the server is processed according to a security protocol. In case it is determined that the overload protection processing has been requested, the message is processed according to the security protocol by adopting the overload protection processing, and a predetermined indication is included in the message processed according to the security protocol by adopting the overload protection processing, in case the message is not dropped.

A method for operating an Ethernet communication device having multiple external physical interfaces for a motor vehicle includes the following steps: detecting a special state on at least one of the interfaces, generating a control signal to adapt the interface in the special state if the special state is detected; and blocking a forwarding of a message packet arriving on the interface in the special state to a media access control unit of the Ethernet communication device on the basis of the control signal.

In a load sharing method and a router device, each of the router devices in a load sharing relationship obtains load sharing information of all downstream router devices that are in a load sharing relationship with the router device, and when determining load sharing information of the router device, the router device determines the load sharing information of the router device according to the load sharing information of all the downstream router devices that are in a load sharing relationship with the router device, so that a disturbance factor of the router device is different from disturbance factors of all the downstream router devices that are in a load sharing relationship with the router device. Therefore, a coupling degree of load sharing information of all the router devices in a load sharing relationship in the network is reduced, and load sharing at all levels is even.

Disclosed is a time-division multiplexing (TDM) scheduler capable of determining a service order for serving N packet transmission requesters. The TDM scheduler includes: N current count value generators configured to serve the N packet transmission requesters respectively, and generate N current count values according to parameters of the N packet transmission requesters, a previous scheduling result generated by the EDD scheduler previously, and a predetermined counting rule; and an earliest due date (EDD) scheduler configured to generate a current scheduling result for determining the service order according to the N current count values and a predetermined urgency decision rule, wherein an extremum of the N current count values relates to one of the N packet transmission requesters, and the EDD scheduler selects this requester as the one to be served preferentially.

A method includes measuring input/output traffic for respective hosts that are connected to a Fibre Channel N_Port Virtualizer (FC-NPV) switch, which is in communication with a first N_Port ID Virtualization (NPIV) core switch via a first port channel and with a second NPIV core switch via a second port channel; determining that traffic carried on the first port channel between the FC-NPV switch and the first NPIV Core switch exceeds a predetermined threshold compared to traffic carried on the second port channel; and re-assigning traffic from a given host carried on the first port channel to the second port channel between the FC-NPV switch and the second NPIV core switch.