Abstract: Techniques are disclosed describing an In-Service Software Upgrade for a Software Defined Networking (SDN) controller of a cloud data center. A cloud data center includes a first SDN controller configured to perform session management of a plurality of session instances, an orchestration engine communicatively coupled to the first SDN controller via a northbound application program interface (API); and a plurality of compute nodes configured to perform compute functions for the plurality of session instances. In response to receiving a request to perform an in-service software upgrade (ISSU) of the first SDN controller, the orchestration engine spawns a second, upgraded SDN controller on the cloud data center. An ISSU process initializes the second SDN controller and transfers session management for the plurality of session instances from the first SDN controller to the second SDN controller. Finally, the ISSU process removes the first SDN controller from the cloud data center.

Abstract: Techniques are described to provide multicast service within a virtual network using a virtual network controller and endpoint replication without requiring multicast support in the underlying network. The virtual network controller is configured to create a multicast tree for endpoint devices of a multicast group in the virtual network at a centralized location instead of in a distributed fashion. The virtual network controller communicates the multicast tree to one or more of the endpoint devices of the multicast group to instruct the endpoint devices to replicate and forward multicast packets to other endpoint devices according to the multicast tree. The replication and forwarding of multicast packets is performed by virtual switches executed on the endpoint devices in the virtual network. No replication is performed within the underlying network. The techniques enable multicast service within a virtual network without requiring multicast support in the underlying network.

Abstract: In general, techniques are described for configuring and managing virtual networks. For example, a distributed virtual network controller is described that configures and manages an overlay network within a physical network formed by plurality of switches. A plurality of servers are interconnected by the switch fabric, each of the servers comprising an operating environment executing one or more virtual machines in communication via the overlay networks. The servers comprises a set of virtual switches that extends the overlay network as a virtual network to the operating environment of the virtual machines.

Abstract: In general, techniques are described for facilitating multi-tenancy of a server accessed by virtual networks of a data center. A device included within a data center comprising one or more processors may perform the techniques. The processors may be configured to execute a virtual switch that supports a number of virtual networks executing within the data center. The virtual switch may be configured to receive a request regarding data associated with an identifier that is unique within one of the virtual networks that originated the request. The virtual switch may then translate the identifier included within the request to generate a globally unique identifier that is unique within the plurality of virtual networks, update the request to replace the identifier included within the request with the globally unique identifier, and transmit the updated request to a server of the data center.

Abstract: In one example, a network device includes a virtual network agent, and a network interface to send network packets to the virtual network controller using a default route for a physical network prior to establishing a communication session between a virtual network controller and the virtual network agent, wherein, after establishing the communication session between the virtual network controller device and the virtual network agent, the virtual network agent receives from the virtual network controller a command to install a new route at the network device, wherein the new route specifies encapsulation information to use for encapsulating network packets for sending the network packets to the virtual network controller over an overlay network, and wherein, responsive to detecting a failed link in the physical network, the virtual network agent sends packets to the virtual network controller on an alternate route in the overlay network.

Abstract: In one example, a network device includes a virtual network agent, and a network interface to send network packets to the virtual network controller using a default route for a physical network prior to establishing a communication session between a virtual network controller and the virtual network agent, wherein, after establishing the communication session between the virtual network controller device and the virtual network agent, the virtual network agent receives from the virtual network controller a command to install a new route at the network device, wherein the new route specifies encapsulation information to use for encapsulating network packets for sending the network packets to the virtual network controller over an overlay network, and wherein, responsive to detecting a failed link in the physical network, the virtual network agent sends packets to the virtual network controller on an alternate route in the overlay network.

Abstract: Techniques are described to provide multicast service within a virtual network using a virtual network controller and endpoint replication without requiring multicast support in the underlying network. The virtual network controller is configured to create a multicast tree for endpoint devices of a multicast group in the virtual network at a centralized location instead of in a distributed fashion. The virtual network controller communicates the multicast tree to one or more of the endpoint devices of the multicast group to instruct the endpoint devices to replicate and forward multicast packets to other endpoint devices according to the multicast tree. The replication and forwarding of multicast packets is performed by virtual switches executed on the endpoint devices in the virtual network. No replication is performed within the underlying network. The techniques enable multicast service within a virtual network without requiring multicast support in the underlying network.

Abstract: In general, techniques are described for facilitating multi-tenancy of a server accessed by virtual networks of a data center. A device included within a data center comprising one or more processors may perform the techniques. The processors may be configured to execute a virtual switch that supports a number of virtual networks executing within the data center. The virtual switch may be configured to receive a request regarding data associated with an identifier that is unique within one of the virtual networks that originated the request. The virtual switch may then translate the identifier included within the request to generate a globally unique identifier that is unique within the plurality of virtual networks, update the request to replace the identifier included within the request with the globally unique identifier, and transmit the updated request to a server of the data center.

Abstract: A system and method for improving the speed of packet switching in a routing platform that maps shared I/O memory into two address spaces. The first address space is mapped with the cache attribute and uses the cache write through attribute. Addresses in this address space are not equal to the physical address in the shared I/O memory and are translated to the physical addresses. Code executed by the CPU to switch packets utilizes the first address space to access packet data. The second address space is mapped with the non-cache attribute and addresses in this space are equal to the physical addresses in the shared I/O memory. The second address space is utilized by I/O devices when accessing shared I/O memory. Addresses of buffers for storing packet data in the shared I/O memory are converted from the first address space to the second address space when given to I/O devices.