Avago Technologies International Sales Pte. Limited

Brocade Communications Systems, Inc.

Cisco Technology, Inc.

Nikunj Vaidya

Eswara S P Chinthalapati

Clarence Filsfils

Pradosh Mohapatra

Vardarajan Venkatesh

Pankaj Bhagra

Arun Satyanarayana

Scott Crowns

Eswara SP Chinthalapati

One embodiment of the present invention provides a switch in a software-defined network. The switch includes at least one port, a flow management module, and forwarding circuitry. The port is capable of receiving a frame belonging to a software-defined data flow and a frame belonging to a regular data flow. The flow management module logically partitions the port for the frame belonging to the software-defined data flow from the frame belonging to the regular data flow. The forwarding circuitry forwards the frame belonging to the software-defined data flow based on a flow definition in a local flow table. The flow definition indicates how the software-defined data flow is processed in a software-defined network.

Integrated heterogeneous software-defined network

That each switch is configured with a series of new commands, such as NELD, which include various parameters, the parameters matching predefined terms, such as STP, BGP or DROP, and specifying a particular port or interface, if relevant. By predefining the parameters, the switch is configured to obtain data relevant to a particular problem. When a problem occurs, the administrator sends the command to all relevant switches with a parameter relevant to the problem being analyzed. Preferably, this sending of the command is done using a management application, so that the command can be sent to each switch by the management application in an extremely short period. By capturing all of the data in a small window, the odds are greatly improved on obtaining the relevant data. By having the preprogrammed commands, all of the desired data can be obtained in a single command rather than a series of commands.

Network event locus descriptor

Improved debugging capabilities for network packet path tracing. Embodiments trace both the control and data planes. During control plane operations each switch appends its identity to the payload, providing a full trace of the control plan path. Responses containing the forward path payload are provided back at each hop, the responses being routing back by tracing back the forward direction control plane. The data plane is monitored by setting traps along the control plane path, with responses at each hop that indicate a given switch has been used being returned along the control plane path.

Packet tracing through control and data plane operations

In one embodiment, a network device may detect a data plane critical fault condition, while a corresponding control plane is not experiencing a critical fault condition. In response to a network device based critical fault condition, the network device may activate and advertise an increased and expensive usable metric for each network interface of the network device. On the other hand, in response to an interface based critical fault condition, the network device may activate and advertise an increased and expensive usable metric for one or more particular network interfaces of the interface based critical fault, and signals, over the control plane to a corresponding network device at an opposing end of each particular network interface of the interface based critical fault, a request to activate and advertise an increased and expensive usable metric at the opposing end of each particular network interface.

Graceful handling of critical traffic blackholing faults

Growing community of inventors

Milpitas, CA, United States of America

Nikunj Vaidya