Abstract: Proposed herein at techniques for scalable network traffic steering in a multi-tenant network. In one aspect, a method includes receiving, at an ingress of a multi-tenant network, a data packet, the data packet including a Virtual Network Identifier (VNI) identifying a corresponding tenant for the data packet in the multi-tenant network, and determining, at the ingress, a corresponding cluster of routers for processing the data packet based on the VNI, wherein the corresponding cluster of routers is one of a plurality of clusters of routers in the multi-tenant network. The method further includes forwarding the data packet to a router in the corresponding cluster of routers for processing.

Abstract: The present technology pertains to receiving, by a multi-tenanted cloud-native headend, network traffic from a source that is directed to a destination within a network of a first tenant, where the network traffic is routed using BGP over a multi-tenanted BGP network, where the multi-tenanted BGP network utilizes route isolation to isolate the network of the first tenant from networks of other tenants making use of the multi-tenanted BGP network, mapping the received network traffic based on at least a source IP, source port, and virtual network interface (VNI) to a port associated with the network of the first tenant over the isolated BGP route, where the mapping is a stateful mapping, where the mapping is also potentially based on other information associated with the network traffic, and directing the received network traffic to a local address in the network of the first tenant.

Abstract: Disclosed is a method for load balancing at the source side rather than the destination side of a network. Destination networks can communicate with source load balancers by transmitting network protocol advertisements such as border gateway protocol (BGP) advertisements. The advertisements can communicate prefixes representing available subnets for load balancing purposes. The source load balancer would then load balance and transmit the network traffic along a path made up at least in part by the prefixes advertised by the network.

Abstract: The present technology pertains to receiving, by a multi-tenanted cloud-native headend, network traffic from a source that is directed to a destination within a network of a first tenant, where the network traffic is routed using BGP over a multi-tenanted BGP network, where the multi-tenanted BGP network utilizes route isolation to isolate the network of the first tenant from networks of other tenants making use of the multi-tenanted BGP network, mapping the received network traffic based on at least a source IP, source port, and virtual network interface (VNI) to a port associated with the network of the first tenant over the isolated BGP route, where the mapping is a stateful mapping, where the mapping is also potentially based on other information associated with the network traffic, and directing the received network traffic to a local address in the network of the first tenant.

Abstract: Proposed herein at techniques for scalable network traffic steering in a multi-tenant network. In one aspect, a method includes receiving, at an ingress of a multi-tenant network, a data packet, the data packet including a Virtual Network Identifier (VNI) identifying a corresponding tenant for the data packet in the multi-tenant network, and determining, at the ingress, a corresponding cluster of routers for processing the data packet based on the VNI, wherein the corresponding cluster of routers is one of a plurality of clusters of routers in the multi-tenant network. The method further includes forwarding the data packet to a router in the corresponding cluster of routers for processing.

Abstract: Techniques for a hub node, provisioned in a site of a hub and spoke overlay network, to receive, store, and/or forward network routing information associated with a spoke, and send packets directly to spoke(s) that are remote from the hub node. A first hub node may receive a network advertisement including a border gateway protocol (BGP) large community string from a first spoke local to the first hub node. The first hub node may send the BGP large community string to a second hub node remote from the first hub node. The second hub node may decode network routing information from the BGP large community string and store the network routing information locally. The second hub node may send a packet from a second spoke local to the second hub node directly to the first spoke without the data packet being routed via the first hub node.

Abstract: Techniques for a hub node, provisioned in a site of a hub and spoke overlay network, to receive, store, and/or forward network routing information associated with a spoke, and send packets directly to spoke(s) that are remote from the hub node. A first hub node may receive a network advertisement including a border gateway protocol (BGP) large community string from a first spoke local to the first hub node. The first hub node may send the BGP large community string to a second hub node remote from the first hub node. The second hub node may decode network routing information from the BGP large community string and store the network routing information locally. The second hub node may send a packet from a second spoke local to the second hub node directly to the first spoke without the data packet being routed via the first hub node.