An embodiment detects, by a packet detection module on a first worker node, a first synchronization message (SYN) issued from a client pod on the first worker node and directed to a first server pod on a second worker node and generates a first SYN record in a connection table stored on the first worker node. The embodiment detects, by the packet detection module, a second SYN issued from the client pod and directed to the first server pod and generates a server pod status record stored in memory on the first worker node indicative of the first server pod having an unhealthy status. The embodiment identifies an available server pod having a healthy status and generates a detour routing rule in a routing table on the first worker node indicative of a revised connection in which the identified available server pod replaces the first server pod.

A medium access control circuit includes a processor, N hardware queues, and an interface circuit, where the N hardware queues are divided into a plurality of hardware queue groups. Where the first hardware queue group corresponds to a network property of the data frame based on a first mapping relationship, the first hardware queue corresponds to a service type of the data frame based on a second mapping relationship, the first mapping relationship includes mappings from network properties to hardware queue groups, and the second mapping relationship includes mappings from a plurality of service types to a plurality of hardware queues in the hardware queue group corresponding to the network property of the data frame; and then, the interface circuit sends the data frame from the N hardware queues through a radio channel.

A medium access control circuit includes a processor, N hardware queues, and an interface circuit, where the N hardware queues are divided into a plurality of hardware queue groups. Where the first hardware queue group corresponds to a network property of the data frame based on a first mapping relationship, the first hardware queue corresponds to a service type of the data frame based on a second mapping relationship, the first mapping relationship includes mappings from network properties to hardware queue groups, and the second mapping relationship includes mappings from a plurality of service types to a plurality of hardware queues in the hardware queue group corresponding to the network property of the data frame; and then, the interface circuit sends the data frame from the N hardware queues through a radio channel.

A TOE component obtains a first storage address, where the first storage address is an address of a first storage block in a memory, the first storage block stores a target TCP packet, and the target TCP packet includes a packet header and a TCP payload. The TOE component obtains the packet header from the first storage block based on the first storage address. The TOE component performs TCP-related protocol processing based on the packet header, where the TCP payload is not read out of the first storage block by the TOE component when the TOE component performs TCP-related protocol processing based on the packet header.

The disclosed technology teaches testing a mesh network using new service application level KPIs that extend the TWAMP measurement architecture. A control-client receives and parses a configuration file to populate memory with IP addresses, ports, and test session parameters for disclosed KPIs used to originate two-way test sessions from a first network host; with control-servers and session-reflectors. The method extends the receiving, parsing and originating to dozens to thousands of control-clients, by sending to the control-clients configuration files to originate respective test sessions with control-servers in a mesh network using respective test session parameters; and while the test is running, sending an updated configuration file to at least one control-client that introduces a new control-server or replaces a control-server; and expanding the test to include the new or replacement control-server without stopping or restarting TW test sessions with other control-servers; and monitoring the running test sessions and receiving results.

The disclosed technology teaches testing a mesh network using new service application level KPIs that extend the TWAMP measurement architecture. A control-client receives and parses a configuration file to populate memory with IP addresses, ports, and test session parameters for disclosed KPIs used to originate two-way test sessions from a first network host; with control-servers and session-reflectors. The method extends the receiving, parsing and originating to dozens to thousands of control-clients, by sending to the control-clients configuration files to originate respective test sessions with control-servers in a mesh network using respective test session parameters; and while the test is running, sending an updated configuration file to at least one control-client that introduces a new control-server or replaces a control-server; and expanding the test to include the new or replacement control-server without stopping or restarting TW test sessions with other control-servers; and monitoring the running test sessions and receiving results.

In some examples, universal host and NVMe™ storage domain discovery for NVM Express™ over Fabrics (NVMe-oF™) may include broadcasting location parameters to a host and a plurality of NVMe™ storage domains for discovery of a NVM Express™ over Fabrics (NVMe-oF™) service. Based on host parameters and NVMe™ storage domain parameters received in response to the broadcast location parameters, the host and the NVMe™ storage domains may be respectively registered with the NVMe-oF™ service. A mapping that indicates a specified NVMe™ storage domain that is to communicate with the host may be obtained. Communication may be implemented between the host and the specified NVMe™ storage domain by forwarding the mapping and associated discovery target parameters to the host to implement discovery of the specified NVMe™ storage domain by the host, and connection of the specified NVMe™ storage domain to the host.

In some examples, universal host and NVMe™ storage domain discovery for NVM Express™ over Fabrics (NVMe-oF™) may include broadcasting location parameters to a host and a plurality of NVMe™ storage domains for discovery of a NVM Express™ over Fabrics (NVMe-oF™) service. Based on host parameters and NVMe™ storage domain parameters received in response to the broadcast location parameters, the host and the NVMe™ storage domains may be respectively registered with the NVMe-oF™ service. A mapping that indicates a specified NVMe™ storage domain that is to communicate with the host may be obtained. Communication may be implemented between the host and the specified NVMe™ storage domain by forwarding the mapping and associated discovery target parameters to the host to implement discovery of the specified NVMe™ storage domain by the host, and connection of the specified NVMe™ storage domain to the host.

A network device includes a memory and a processor. The processor implements at least one classifier to provide classification for tunneled IPinIP traffic based on classification parameters provided by Internet Protocol Version 6 (IPv6) classification encoding, Internet Protocol Version 4 (IPv4) classification encoding, and Transport Control Protocol/User Datagram Protocol (TCP/UDP) classification encodings, wherein one of the IPv6 classification encoding and the IPv4 classification encoding is provided by extension information to provide extended classification capabilities. The processor uses the at least one classifier to compare the classification parameters provided by the IPv6 classification encoding, the IPv4 classification encoding, and the TCP/UDP classification encodings to fields in an IPv6 header, an IPv4 header, and a TCP/UDP header of the IPinIP packet. Based on the comparison, the processor maps the IPinIP packet to a service flow identified by the classification parameters.

Methods and systems for an accelerated and offload dual border relay. A method includes receiving, by a hardware border relay from a network device, an Internet Protocol (IP) packet, determining, by the hardware border relay, a packet type of the IP packet, translating, by the hardware border relay provisioned with IPv6 transition technology rules, the IP packet to a hardware translated IP packet when the IP packet is a first type, translating, by the offload border relay provisioned with MAP-T rules, the IP packet to an offload translated IP packet when the IP packet is a second type, transmitting, by the offload border relay to the hardware border relay, the offload translated IP packet when the IP packet is the second type, and transmitting, by the hardware border relay, one of the offload translated IP packet and the hardware translated IP packet to another network device.