Abstract: A networking system may include a switch coupled to a computing resource. A resource management system may control the computing resource. A controller may be coupled to the switch. The controller may include a resource management interface that is coupled to the resource management system via a communications link. The resource management interface may receive computing resource information for the computing resource via the communications link. The controller may provide control data to the switch to update a cloud network for the computing resource based on the received computing resource information.

Abstract: A packet forwarding network may include spine and leaf switches that forward network traffic between end hosts. The packet forwarding network may be implemented on multiple network racks in a rack-based system. A controller may control the underlying spine and leaf switches to form on-premise virtual private cloud (VPC) resources. In particular, the controller may form enterprise VPC (EVPC) tenants, each having a virtual router that performs routing between different segments within the corresponding EVPC tenant. The different segments may separately include web, application, and database servers, as end hosts. The controller may form a system VPC tenant having a virtual system router that performs routing between different EVPC tenants. A segment in an internal VPC tenant formed by the controller and/or an external VPC tenant formed by the controller may provide external network access for one or more of the EVPC tenants.

Abstract: Systems and techniques are described for configuring path selection in a network. The network can comprise a first router, a second router, a third router, a fourth router, and an intermediary device. The second router can be configured to use Differentiated Services Code Point (DSCP) while routing packets so that packets with a first DSCP value are routed through the third router, and packets with a second DSCP value are routed through the fourth router. The intermediary device can be configured to: (1) transparently intercept a packet forwarded by the first router to the second router, (2) determine whether the packet is to be routed through the third router or the fourth router, (3) modify a DSCP field in the packet based on said determining, and (4) forward the packet to the second router.

Abstract: A packet forwarding network may include spine and leaf switches that forward network traffic between end hosts. The packet forwarding network may be implemented on multiple network racks in a rack-based system. A controller may control the underlying spine and leaf switches to form on-premise virtual private cloud (VPC) resources. In particular, the controller may form enterprise VPC (EVPC) tenants, each having a virtual router that performs routing between different segments within the corresponding EVPC tenant. The different segments may separately include web, application, and database servers, as end hosts. The controller may form a system VPC tenant having a virtual system router that performs routing between different EVPC tenants. A segment in an internal VPC tenant formed by the controller and/or an external VPC tenant formed by the controller may provide external network access for one or more of the EVPC tenants.

Abstract: A networking system may include a switch coupled to a computing resource. A resource management system may control the computing resource. A controller may be coupled to the switch. The controller may include a resource management interface that is coupled to the resource management system via a communications link. The resource management interface may receive computing resource information for the computing resource via the communications link. The controller may provide control data to the switch to update a cloud network for the computing resource based on the received computing resource information.

Abstract: A packet forwarding network may include spine and leaf switches that forward network traffic between end hosts that are coupled to the packet forwarding network. A controller may control the switches in the forwarding network to implement desired forwarding paths. The packet forwarding network may additionally include access switches that are directly connected to edge switches to form access switch groups. The access switches may also be controlled by the controller. End hosts may be coupled to access switches as well as leaf switches to extend the reach of the packet forwarding network. Additionally, the access switches may include access-leaf switches and access-edge switches to improve network routing capabilities and network flexibility.

Abstract: A packet forwarding network may include spine and leaf switches that forward network traffic between end hosts that are coupled to the packet forwarding network. A controller may control the switches in the forwarding network to implement desired forwarding paths. The packet forwarding network may additionally include access switches that are directly connected to edge switches to form access switch groups. The access switches may also be controlled by the controller. End hosts may be coupled to access switches as well as leaf switches to extend the reach of the packet forwarding network. Additionally, the access switches may include access-leaf switches and access-edge switches to improve network routing capabilities and network flexibility.

Abstract: Methods, systems, and computer programs are presented for distributing network address translation (NAT) operations to a plurality of network devices on a network. One method includes an operation for identifying, by a controller that controls a network fabric, a plurality of switches in the network fabric, each switch having a module for NAT and being configured to forward packets received at the switch. The controller identifies hosts having at least one internal Internet Protocol (IP) address, and for each of the hosts, the controller selects one of the switches from the plurality of switches for performing the NAT for the host. Further, the controller configures the network fabric to cause the selected switch to perform the NAT for the host to enable the host to communicate with an external network. In case of switch failure, the system reallocates NAT loads to other switches for high availability.

Abstract: Systems and techniques are described for configuring path selection in a network. The network can comprise a first router, a second router, a third router, a fourth router, and an intermediary device. The second router can be configured to use Differentiated Services Code Point (DSCP) while routing packets so that packets with a first DSCP value are routed through the third router, and packets with a second DSCP value are routed through the fourth router. The intermediary device can be configured to: (1) transparently intercept a packet forwarded by the first router to the second router, (2) determine whether the packet is to be routed through the third router or the fourth router, (3) modify a DSCP field in the packet based on said determining, and (4) forward the packet to the second router.

Abstract: Systems and techniques are described for path selection. A packet can be transparently intercepted at an intermediary device. Next, the intermediary device may modify one or more bits in the header of the packet. The intermediary device can then forward the packet to the next hop device. In some network configurations, the modifications to the one or more bits in the header of the packet may cause a downstream device to select a path that is different from the path that would have been selected by the downstream device if the one or more bits in the header of the packet had not been modified. A path selection policy can be used to determine whether or not one or more bits in the header of the packet are to be modified.

Abstract: Methods, systems, and computer programs are presented for distributing network address translation (NAT) operations to a plurality of network devices on a network. One method includes an operation for identifying, by a controller that controls a network fabric, a plurality of switches in the network fabric, each switch having a module for NAT and being configured to forward packets received at the switch. The controller identifies hosts having at least one internal Internet Protocol (IP) address, and for each of the hosts, the controller selects one of the switches from the plurality of switches for performing the NAT for the host. Further, the controller configures the network fabric to cause the selected switch to perform the NAT for the host to enable the host to communicate with an external network. In case of switch failure, the system reallocates NAT loads to other switches for high availability.

Abstract: Systems and techniques are described for path selection. A packet can be transparently intercepted at an intermediary device. Next, the intermediary device may modify one or more bits in the header of the packet. The intermediary device can then forward the packet to the next hop device. In some network configurations, the modifications to the one or more bits in the header of the packet may cause a downstream device to select a path that is different from the path that would have been selected by the downstream device if the one or more bits in the header of the packet had not been modified. A path selection policy can be used to determine whether or not one or more bits in the header of the packet are to be modified.

Abstract: Systems and techniques are described for path selection. A packet can be transparently intercepted at an intermediary device. Next, the intermediary device may modify one or more bits in the header of the packet. The intermediary device can then forward the packet to the next hop device. In some network configurations, the modifications to the one or more bits in the header of the packet may cause a downstream device to select a path that is different from the path that would have been selected by the downstream device if the one or more bits in the header of the packet had not been modified. A path selection policy can be used to determine whether or not one or more bits in the header of the packet are to be modified.

Abstract: Systems and techniques are described for path selection. A packet can be transparently intercepted at an intermediary device. Next, the intermediary device may modify one or more bits in the header of the packet. The intermediary device can then forward the packet to the next hop device. In some network configurations, the modifications to the one or more bits in the header of the packet may cause a downstream device to select a path that is different from the path that would have been selected by the downstream device if the one or more bits in the header of the packet had not been modified. A path selection policy can be used to determine whether or not one or more bits in the header of the packet are to be modified.