Abstract: Techniques are provided for optimizing multicast routing in a network. At a router device, a message is sent to one or more physical devices. The message is configured to solicit a response indicating a network assignment for each of the physical devices. A response message is received from each of the physical devices. The response message comprises network assignment information for each of the physical devices. For each of the physical devices, the network assignment information is translated into a segment identifier. The segment identifier is distributed to other router devices in the network.

Abstract: In one embodiment, a method includes storing a service topology route at a network device interconnecting at least two zones comprising a plurality of hosts, and propagating the service topology route to create a service chain comprising a service node in communication with the network device. The service topology route creates a forwarding state at network devices in the service chain for use in inter-zone routing in a virtual private network. An apparatus and logic are also disclosed herein.

Abstract: Techniques are presented herein for optimizing traffic routing in overlay networks. At a first edge device located at a first site in a network, a message is received that indicates address information of a network device. The address information of the network device is stored in an address table. The address information is associated with a site identifier that identifies a second site at which the network device is located. The site identifier is mapped to an identifier associated with a second edge device that is responsible for routing traffic to network devices at the second site.

Abstract: A first provider edge network device that is configured in a multi-homed virtual private network for a data center in which there are one or more peer edge network devices including a second edge network device, receives from the second edge network device a message indicating that a link for a particular Ethernet segment of the second edge network device in the data center is down. Information is stored at the first edge network device indicating state of links for Ethernet segments associated with each of the one or more other edge network devices at the data center. The first edge network device forwards of traffic for the particular Ethernet segment locally on Ethernet segments in the data center on behalf of the second edge network device. The proxy forwarding is performed for traffic for the particular Ethernet segment that originates from the data center, that is, for “same-site” traffic.

Abstract: In one embodiment, a method includes building an address resolution cache for a layer-3 router in a first layer-2 datacenter, where the address resolution cache includes a plurality of entries, each of the entries containing a host network address, a host hardware address, and a switch identifier for a switch serving a host, intercepting an address resolution flood within the first layer-2 datacenter that seeks address resolution for a host in a second layer-2 datacenter, and generating a response to the address resolution flood that indicates a source in the second layer-2 datacenter, where data indicating the source in the second layer-2 datacenter is accessed from the address resolution cache.

Abstract: In one embodiment, a method includes receiving a packet at an edge device in a first network site in communication with a second network site through a transport network, the packet comprising a destination address for a host at the second network site, verifying at the edge device a connection with the host, and inserting the destination address in a forwarding information base at the edge device upon verifying the connection with the host. An apparatus and logic are also disclosed herein.

Abstract: Presented herein are techniques to handle data packets received at a top-of-rack (ToR) switch in an underlay network hosting one or more tenant networks. The underlay network may comprise a plurality of ToR switches each connected to each of a plurality of spine switches. The data packet may be received from a virtual machine in a tenant network, and have a header with a data packet identifier. A mobility domain identifier may be determined that corresponds to the network portion within which the virtual machine can be migrated. The mobility domain may be configured on the ToR switch on a per-port basis based on the virtual machine connected to the ToR switch port. A unique identifier may be formed based on the data packet identifier and mobility domain identifier. Using this unique identifier, provisioning information may be obtained for the tenant network on the ToR switch.

Abstract: Techniques are provided for optimizing multicast routing in a network. At a router device, a message is sent to one or more physical devices. The message is configured to solicit a response indicating a network assignment for each of the physical devices. A response message is received from each of the physical devices. The response message comprises network assignment information for each of the physical devices. For each of the physical devices, the network assignment information is translated into a segment identifier. The segment identifier is distributed to other router devices in the network.

Abstract: Devices, methods and instructions encoded on computer readable medium are provided herein for creation of an overlay network on a non-multicast or source specific multicast (SSM) core. In one example, virtual private network (VPN) adjacencies are established between an adjacency server and one or more edge devices each located at different network sites. A unicast replication list is then generated at the adjacency server. The unicast replication list includes the Internet Protocol addresses for each of the edge devices having VPN adjacencies with the adjacency server. The unicast replication list is then advertised to each of the edge devices for use in establishing VPN adjacencies with one another.

Abstract: An example method for bi-directional flow stickiness in a network environment is provided and includes receiving a packet of a flow from a source address in a network, and destined to a destination address outside the network, looking up a flow table for an entry corresponding to the flow, and caching a forward flow entry and a reverse flow entry in the flow table if the look up results in a miss, the forward flow entry indicating a routing lookup, and the reverse flow entry indicating a previous hop from which the packet was received to maintain a bi-directional flow stickiness. Some embodiments also include receiving another packet from the destination address outside the network destined to the source address in the network, and forwarding it to the previous hop listed in the reverse flow entry.

Abstract: Techniques provided herein enable a set of independent, unconnected devices in a network to support distributed Layer 3 (L3) gateway functionality for an overlay based virtual network by intelligently triggering proxy addressing information updates.

Abstract: A method and apparatus to provide network-wide enhanced load balancing. In one embodiment a first message comprising a first value and an address is generated. The first message is transmitted to a first node via a first communication link. A second message comprising a second value and the address is generated. The first and second values are unequal. The second message is transmitted to the first node via a second communication link. Packets are received, wherein each of the packets comprises the address. A first percentage of the packets are forwarded to a first device via the first communication link, wherein the first percentage is proportional to the first value. A second percentage of the packets are forwarded to a second device via the second communication link, wherein the second percentage is proportional to the second value.

Abstract: According to one embodiment, a method includes receiving, by a first edge device at a first site, a first site overlay control plane message including control plane information. The first edge device translates the first site overlay control plane message into a core overlay control plane message. The first edge device sends the core overlay control plane message over a core network to a second edge device at a second site.

Abstract: In one embodiment, an autonomous system border router (ASBR) advertises a same forwarding label for received advertised routes of a merging context that were advertised with a same forwarding label for the ASBR to use when sending corresponding packets. An ASBR receives via a routing protocol from a particular router in the same autonomous system, a plurality of same-labeled received routes advertised with a same first forwarding label within a merging context. In response to each of the plurality of same-labeled received routes having the same first forwarding label to use to forward packets to the particular router and being in the same merging context, the ASBR determines a merged forwarding label and advertises to a peer ASBR in another autonomous system (AS) each of the plurality of same-labeled received routes with the merged forwarding label for the peer ASBR to use to forward packets to the ASBR.

Abstract: Devices, methods and instructions encoded on computer readable medium for implementation of a dual-adjacency between edge devices of a network site. A first edge device comprises one or more local interfaces configured for communication, via a local network, with one or more network devices co-located in a first network site. The first edge device also comprises one or more overlay interfaces configured for communication, via a core network, with one or more network devices located in one or more other network sites connected to the core network. The first edge device comprises a processor configured to establish, via at least one of the local interfaces, a site communication channel with a second edge device co-located in the first network site. The processor is further configured to establish an overlay communication channel, via at least one of the overlay interfaces, with the second edge device.

Abstract: In one embodiment, a method at a network device includes receiving a link layer advertisement, comparing information in the link layer advertisement with connectivity information stored at the network device, and based on the comparison, determining if there is a cabling error between the network device and a link peer transmitting the link layer advertisement. An apparatus and logic are also disclosed herein.

Abstract: Devices, methods and instructions encoded on computer readable medium for implementation of a dual-adjacency between edge devices of a network site. A first edge device comprises one or more local interfaces configured for communication, via a local network, with one or more network devices co-located in a first network site. The first edge device also comprises one or more overlay interfaces configured for communication, via a core network, with one or more network devices located in one or more other network sites connected to the core network. The first edge device comprises a processor configured to establish, via at least one of the local interfaces, a site communication channel with a second edge device co-located in the first network site. The processor is further configured to establish an overlay communication channel, via at least one of the overlay interfaces, with the second edge device.

Abstract: A method of one example embodiment includes receiving at a first network element a packet from a host local to the first network element destined for a remote host; determining that a subnet of the remote host is not instantiated on the first network element; originating a discovery request to discover the remote host, wherein the discovery request is originated in a Virtual Routing Forwarding instance (“VRF”) and identifies the subnet to which the remote host belongs; and broadcasting the discovery request to network elements comprising the VRF. The method may further include, upon receipt of the discovery request, determining whether the identified subnet is configured locally on the second network element and if not, dropping the discovery request; otherwise, rewriting the discovery request to include to an anycast IP address of the remote host's subnet and forwarding the rewritten request.

Abstract: An example method is provided in one example embodiment and includes receiving first values associated with a host located in a first overlay domain; translating the first values to second values, the second values being local values in a second overlay domain; storing the second values and corresponding first values in a mapping table for the second overlay domain; and advertising the second values for the host in the second overlay domain.

Abstract: Presented herein are techniques to detect a misconfigured duplicate Internet Protocol (IP) address in a distributed data center network fabric. A network topology is provided that includes a plurality of network elements. A network element receives a configuration message comprising an IP address from a first host associated with the first network element. The IP address is detected to already be in use by a second host associated with a second network element. A request message is provided to the second network element, the request message configured to cause the second network element to attempt to contact the second host. A reply message is received from the second network element. If the reply message indicates that the second host exists at the second network element, an indication of a misconfiguration event is generated. Otherwise, the IP address is associated with the first host.