Nicira, Inc.

Vmware, Inc.

Juniper Networks, Inc.

Xinhua Hong

Yong Wang

Jayant Jain

Anirban Sengupta

Jia Yu

Ronghua Zhang

Mohan Parthasarathy

Teemu Koponen

Dexiang Wang

Kai-Wei Fan

Mike Parsa

David J Leroy

Raju Koganty

Rick Lund

Michael M Hu

Krishna Narayanaswamy

Ying Zhang

Hongbin Wang

Subrahmanyam Manuguri

Sreeram Kumar Ravinoothala

Stephen Tan

Anil Lohiya

Senthilkumar Karunakaran

Awan Kumar Sharma

Meenakshi Sundaram Selvaraj

Vijayalaxmi Basavaraj

Yu Ying

Xinghua Hu

Ganesan Chandrashekhar

Serge Maskalik

Abhishek Goliya

Jochen Behrens

Sharath Bhat

Michael Luo

Wenyi Jiang

Guolin Yang

Hongwei Zhu

Bhargav Puvvada

Eduard Serra Miralles

Liwen Wu

Rakesh Nair

Mochi Xue

Henry Han

Pankaj Gupta

Some embodiments provide a method for a data message processing device that includes multiple network interfaces associated with at least two different non-uniform memory access (NUMA) nodes. The method receives a data message at a first network interface associated with a particular one of the NUMA nodes. Based on processing of the data message, the method identifies multiple output options for the data message. Each of the output options has an equal forwarding cost and each output option is associated with a respective one of the NUMA nodes. The method selects an output option for the data message that is associated with the particular NUMA node to avoid cross-NUMA node processing of the data message.

NUMA-aware output interface selection

A novel method for stateful packet classification that uses hardware resources for performing stateless lookups and software resources for performing stateful connection flow handshaking is provided. To classify an incoming packet from a network, some embodiments perform stateless look up operations for the incoming packet in hardware and forward the result of the stateless look up to the software. The software in turn uses the result of the stateless look up to perform the stateful connection flow handshaking and to determine the result of the stateful packet classification.

Use of stateless marking to speed up stateful firewall rule processing

In order to enable dynamic scaling of network services at the edge, novel systems and methods are provided to enable addition of add new nodes or removal of existing nodes while retaining the affinity of the flows through the stateful services. The methods provide a cluster of network nodes that can be dynamically resized to handle and process network traffic that utilizes stateful network services. The existing traffic flows through the edge continue to function during and after the changes to membership of the cluster. All nodes in the cluster operate in active-active mode, i.e., they are receiving and processing traffic flows, thereby maximizing the utilization of the available processing power.

Stateful services on stateless clustered edge

A novel design of a gateway that handles traffic in and out of a network by using a datapath pipeline is provided. The datapath pipeline includes multiple stages for performing various data-plane packet-processing operations at the edge of the network. The processing stages include centralized routing stages and distributed routing stages. The processing stages can include service-providing stages such as NAT and firewall. The gateway caches the result previous packet operations and reapplies the result to subsequent packets that meet certain criteria. For packets that do not have applicable or valid result from previous packet processing operations, the gateway datapath daemon executes the pipelined packet processing stages and records a set of data from each stage of the pipeline and synthesizes those data into a cache entry for subsequent packets.

Datapath for multiple tenants

Described herein are systems, methods, and software to manage secure tunnel communications in multi-edge gateway computing environments. In one implementation, a control system identifies an edge gateway from a plurality of edge gateways to support a private network tunnel. The control system further identifies addressing attributes associated with communications directed over the private network tunnel and configures the plurality of edge gateways to forward packets associated with the addressing attributes to the identified edge gateway, wherein the edge gateway can process and forward the packets over the private network tunnel.

Managing exchanges between edge gateways in a cloud environment to support a private network connection

Some embodiments provide, for a gateway datapath that executes on a gateway device to implement tenant logical routers for multiple different tenant logical networks and process traffic between the tenant logical networks and an external network, a method for managing QoS for the plurality of tenant logical networks. The method receives a data message for a particular tenant logical network. The method executes a set of processing stages to process the data message. The set of processing stages includes a processing stage for a particular tenant logical router of the particular tenant logical network. As part of the processing stage for the particular tenant logical router, the method uses a QoS data structure specific to the particular tenant logical router to determine whether to allow the data message. The gateway device stores at least one separate QoS data structure for each of a set of the tenant logical routers.

Tenant-specific quality of service

In some embodiments, a method fragments a first packet into a plurality of fragments when a length of an encapsulated first packet is larger than a maximum transmission unit size. For each fragment in the plurality of fragments, fragmentation information is generated. The method encapsulates each fragment in the plurality of fragments with an outer header to form a plurality of encapsulated packets. The respective fragmentation information for each fragment is inserted in a portion of the outer header that is processed by endpoints of an overlay tunnel and not processed by a device along a path of the overlay tunnel. The plurality of encapsulated packets are sent via the overlay tunnel.

Packet fragmentation using outer header

Some embodiments provide a method for forwarding data messages at multiple edge nodes of a logical network that process data messages between a logical network and an external network. At a particular one of the edge nodes, the method receives a data message sent from a source machine in the logical network. The method performs network address translation to translate a source network address of the data message corresponding to the source machine into an anycast network address that is shared among the edge nodes. The method sends the data message with the anycast network address as a source network address to the external network. Each edge node receives data messages from source machines in the logical network and translates the source addresses of the data messages into the same anycast public network address prior to sending the data messages to the external network.

Anycast address for network address translation at edge

Example methods for a network device to perform address resolution handling. The method may comprise: in response to a first distributed router (DR) port of a first DR instance detecting an address resolution request from a second DR port of a second DR instance, generating a modified address resolution request that is addressed from a first address associated with the first DR port instead of a second address associated with the second DR port. The modified address resolution request may be broadcasted within a logical network that is connected to the first DR instance through network extension. The method may also comprise: in response to detecting an address resolution response that includes protocol-to-hardware address mapping information associated with an endpoint located on the logical network, generating and sending a modified address resolution response towards the second DR port of the second DR instance.

Address resolution handling at logical distributed routers

A novel design of a gateway that handles traffic in and out of a network by using a datapath daemon is provided. The datapath daemon is a run-to-completion process that performs various data-plane packet-processing operations at the edge of the network. The datapath daemon dispatches packets to other processes or processing threads outside of the daemon by utilizing a user space network stack.

Edge datapath using user space network stack

A network system that uses a cluster of edge nodes to send and receive multicast traffic is provided. The network system is a network virtualization environment that includes one or more distributed routers, each distributed router implemented by virtualization software running on one or more host machines. The network system also includes a cluster of edge nodes for sending data from the one or more distributed routers to one or more uplink/upstream physical routers outside of a datacenter and for receiving data from the physical routers to the distributed routers. One of the edge nodes is a designated edge node that queries for membership information for one or more multicast groups to be received by at least two edge nodes of the cluster of edge nodes. The cluster of edge nodes forwards multicast traffic to and from the distributed routers according to the received membership information.

Active-active support of multicast streams in virtualized environment

Described herein are systems, methods, and software to manage the selection of an edge gateway or edge for processing a packet. In one implementation, a first edge may receive a packet and hash addressing information in the packet to select a second edge to process the packet. The first edge may further forward the packet to the second edge, permitting the second edge to process the packet. Once processed, the second edge may forward the packet to a destination host computing system and notify the host computing system to use the second edge for response packets directed at a source internet protocol (IP) address in the packet.

Managing tunnel interface selection between gateways in a computing environment

A novel algorithm for packet classification that is based on a novel search structure for packet classification rules is provided. Addresses from all the containers are merged and maintained in a single Trie. Each entry in the Trie has additional information that can be traced back to the container from where the address originated. This information is used to keep the Trie in sync with the containers when the container definition dynamically changes.

Efficient packet classification for dynamic containers

Some embodiments provide a method for forwarding data messages between edge nodes that perform stateful processing on flows between a logical network and an external network. At a particular edge node, the method receives a data message belonging to a flow. The edge nodes use a deterministic algorithm to select one of the edge nodes to perform processing for each flow. The method identifies a first edge node to perform processing for the flow in a previous configuration and a second edge node to perform processing for the flow in a new configuration according to the algorithm. When the first and second edge nodes are different, the method uses a probabilistic filter and a stateful connection tracker to determine whether the flow existed prior to a particular time. When the flow did not exist prior to that time, the method selects the second edge node for the received data message.

Scaling edge services with minimal disruption

Growing community of inventors

Milpitas, CA, United States of America

Xinhua Hong