Some embodiments provide a method for configuring an edge computing device to implement a logical router belonging to a logical network. The method configures a datapath executing on the edge computing device to use a first routing table associated with the logical router for processing data messages routed to the logical router. The method configures a routing protocol application executing on the edge computing device to (i) use the first routing table for exchanging routes with a network external to the logical network and (ii) use a second routing table for exchanging routes with other edge computing devices that implement the logical router.

Some embodiments provide a method for an edge computing device in a first datacenter that implements a logical network gateway for processing data traffic for a particular LFE between the first datacenter and multiple other datacenters. For each particular other datacenter, the method stores a record that maps logical network addresses for DCNs connected to the particular LFE and operating in the particular datacenter to a group of TEP addresses corresponding to logical network gateways that handle data traffic for the particular LFE between the particular datacenter and the other datacenters, including the first datacenter. Upon receiving a data message for the particular LFE from a host computer in the first datacenter, the method uses a destination address of the data message to identify one of the groups of TEP addresses. The method encapsulates the data message with one of the TEP addresses from the identified group of TEP addresses.

Some embodiments provide a method for an edge computing device in a first datacenter that implements a logical network gateway for processing data traffic for a particular LFE between the first datacenter and multiple other datacenters. For each particular other datacenter, the method stores a record that maps logical network addresses for DCNs connected to the particular LFE and operating in the particular datacenter to a group of TEP addresses corresponding to logical network gateways that handle data traffic for the particular LFE between the particular datacenter and the other datacenters, including the first datacenter. Upon receiving a data message for the particular LFE from a host computer in the first datacenter, the method uses a destination address of the data message to identify one of the groups of TEP addresses. The method encapsulates the data message with one of the TEP addresses from the identified group of TEP addresses.

A programmable network switch includes at least one pipeline including a packet parser configured to parse packets, and a plurality of ports for communication with network devices including a plurality of Data Storage Devices (DSDs). A packet comprising a write command is received to store data in a DSD of the plurality of DSDs, and an identifier generated for the data is compared to a plurality of identifiers generated for data stored in the plurality of DSDs. It is determined whether to send the write command to store the data to the DSD based on whether the generated identifier matches an identifier of the plurality of identifiers. In one aspect, the data to be stored for the write command is extracted from the packet using a pipeline of the programmable network switch, and at least a portion of the extracted data is used to generate the identifier for the data.

A programmable network switch includes at least one pipeline including a packet parser configured to parse packets, and a plurality of ports for communication with network devices including a plurality of Data Storage Devices (DSDs). A packet comprising a write command is received to store data in a DSD of the plurality of DSDs, and an identifier generated for the data is compared to a plurality of identifiers generated for data stored in the plurality of DSDs. It is determined whether to send the write command to store the data to the DSD based on whether the generated identifier matches an identifier of the plurality of identifiers. In one aspect, the data to be stored for the write command is extracted from the packet using a pipeline of the programmable network switch, and at least a portion of the extracted data is used to generate the identifier for the data.

Technologies for hairpinning network traffic include a compute device with a network interface controller (NIC) configured to receive, by a virtual Ethernet port aggregator (VEPA) agent of a media access control (MAC) of the NIC, a network packet from a virtual machine (VM). The VEPA agent is configured to transmit the received network packet to an agent deployed on an accelerator device of the NIC and the agent is configured to forward the received network packet to a virtual Ethernet bridge (VEB) hairpin agent of the accelerator device. The VEB hairpin agent is configured to determine whether a target destination of the network packet corresponds to another VM, return the received network packet to the agent deployed the accelerator device. The agent is further configured to forward the received network packet to the VEPA agent, which is further configured to transmit the received network packet to the other VM.

Embodiments of the present disclosure are directed to protocol state transition and/or resource state transition tracker configured to monitor, e.g., via filters, for certain protocol state transitions/changes or host hardware resource transitions/changes when a host processor in the control plane that performs such monitoring functions is unavailable or overloaded. The filters, in some embodiments, are pre-computed/computed by the host processor and transmitted to the protocol state transition and/or resource state transition tracker. The protocol state transition and/or resource state transition tracker may be used to implement a fast upgrade operation as well as load sharing and or load balancing operation with control plane associated components.

Embodiments of the present disclosure are directed to protocol state transition and/or resource state transition tracker configured to monitor, e.g., via filters, for certain protocol state transitions/changes or host hardware resource transitions/changes when a host processor in the control plane that performs such monitoring functions is unavailable or overloaded. The filters, in some embodiments, are pre-computed/computed by the host processor and transmitted to the protocol state transition and/or resource state transition tracker. The protocol state transition and/or resource state transition tracker may be used to implement a fast upgrade operation as well as load sharing and or load balancing operation with control plane associated components.

Techniques for managing configuration of virtual switches in a virtual machine network are disclosed. In an embodiment, a virtual machine network that includes virtual switches is configured to revert back to a saved network configuration if a configuration change causes the connection between the VM management system and a managed node to be lost. For example, before any configuration changes are made, the active configuration is saved. If the new configuration supports a working connection between the managed node and the VM management system, then the saved configuration is no longer needed and can be flushed from memory. If, however, the new configuration causes the managed node to be disconnected from the VM management system, then the system reverts back to the saved configuration that was previously known to work. The saved configuration is used to reestablish the connection so that the network continues to function.

Some embodiments provide a system for implementing a logical network that spans multiple datacenters. The system includes, at each of the datacenters, a set of host computers that execute (i) data compute nodes (DCNs) belonging to the logical network and (ii) managed forwarding elements (MFEs) that implement the logical network to process data messages for the DCNs executing on the host computers. The system also includes, at each of the datacenters, a set of computing devices implementing logical network gateways for logical forwarding elements (LFEs) of the logical network. The logical network gateways are connected to the logical network gateways for the LFEs at the other datacenters. The MFEs executing on the host computers in a first datacenter communicate with the MFEs executing on the host computers in a second datacenter via the logical network gateways of the first and second datacenters.