Santa Clara, CA, United States
Santa Clara, CA, United States

Arista Networks is a computer networking company headquartered in Santa Clara, California, USA. The company designs and sells multilayer network switches to deliver software-defined networking solutions for large datacenter, cloud computing, high-performance computing and high-frequency trading environments. Arista's products include an array of 10100 Gigabit Ethernet low-latency cut-through switches, including the 7124SX, which remained the fastest switch using SFP+ optics through September 2012, with its sub-500ns latency, as well as the 7500 series, Arista’s award-winning modular 10G100Gbit/s switch. Arista's own Linux-based network operating system, EOS , runs on all Arista products. Wikipedia.


Time filter

Source Type

Patent
Arista Networks | Date: 2016-08-23

Described herein are various embodiments of a network element to process routing entries in a forwarding information base (FIB) for hosts that are directly attached to a multi-layer network element to encode address resolution protocol (ARP) data into forwarding equivalence class (FEC) entries for the directly attached hosts. One embodiment provides for a network element comprising a control plane including a forwarding agent and a data plane coupled to the control plane. The data plane can include a forwarding engine to forward a unit of network data from an ingress interface to an egress interface. The forwarding agent can be configured to program the forwarding engine with a forwarding information base having data-link layer forwarding data encoded into a network-layer forwarding entry.


A method and apparatus of a device that grows and/or shrinks a table that is shared between a writer and a plurality of readers is described. In an exemplary embodiment, a device receives an entry to be added to the shared table. In response to receiving the entry, the device remaps shared table to add a new storage segment to the shared table. The device further adds the entry to the shared table, where the entry is stored in the new storage segment. In addition, the device updates a shared table characteristic to indicate that the shared table has changed. The device further shrinks the shared table by remapping the table to remove a segment of the table.


Patent
Arista Networks | Date: 2016-10-27

A method for virtual extensible local area network (VXLAN) encapsulation. The method includes receiving a first augmented MAC frame on a first ingress port of a first network device, where the first augmented MAC frame includes a first egress port ID (EPID), a first ingress port ID (IPID), and a first MAC frame. The method further includes identifying a first destination VXLAN tunnel endpoint (VTEP) internet protocol (IP) address based on the first EPID, where the first destination VTEP IP address is associated with a first destination VTEP. The method further includes identifying a source VTEP IP address based on the first IPID, performing VXLAN encapsulation of the first MAC frame to obtain a VXLAN frame, and sending the VXLAN frame to the first destination VTEP via a first egress port of the first network device.


A method for initializing the border gateway protocol (BGP) on network devices. The method includes initializing, by a first network device, a BPG session with a second network device. The method further includes receiving, by the first network device from the second network device, route updates and an end of routing information base (EOR) marker. The method further includes updating an n_EOR variable maintained by the first network device after receiving the EOR marker and, after updating the n_EOR variable, determining that the first network device has received route updates from all peer network devices, and subsequently updating, in the first network device, a routing information base (RIB) using route updates received from the peer network devices. The method further includes updating, in the first network device, a forwarding information base (FIB) using the updated RIB and advertising, to all peer network devices by the first network device, routes.


A method for processing IP multicast packets in a MLAG domain. The method includes processing the IP multicast packet using the bridging functionality and the routing functionality implemented by each of the MLAG peers to process the IP multicast packets.


Patent
Arista Networks | Date: 2016-08-29

A system may include a reversible fan module that creates an airflow. The reversible fan module may include a set of attachment members, disposed on a face of the reversible fan module, that secure the reversible fan module a first orientation or a second orientation. The system may further include a chassis that include a divider that divides an interior region of the chassis into a front compartment and a rear compartment, a window, disposed on the divider, that allows airflow between the front compartment and the rear compartment, a first opening in the chassis that opens the front compartment to a first external region, a second opening in the chassis that opens the rear compartment to a second external region, and a bay that receives the reversible fan module.


Patent
Arista Networks | Date: 2016-01-06

Various embodiments of a virtual output queue system within a network element enables per-input port virtual output queues within a network data processor of the network element. In one embodiment, each port managed by a network data processor has an associated set of virtual output queues for each output port on the network data element. In one embodiment, network data processor hardware supports per-processor VOQs and per-input port VOQs are enabled in hardware for layer 3 forwarding by overloading layer 2 forwarding logic. In such embodiment, a mapping table is generated to enable virtual per-input port VOQs for layer 3 forwarding logic using layer 2 logic that is otherwise unused during layer 3 forwarding. In one embodiment, multiple traffic classes can be managed per-input port when using per-input port VOQs. In one embodiment, equal cost multi-path (ECMP) and link aggregation support is also enabled.


A method and apparatus of a device that broadcasts data to multiple hardware forwarding engines is described. In an exemplary embodiment, a central processing unit of the device receives the data to broadcast to the plurality of hardware forwarding engines. The device further writes the data to a broadcast log. In addition, the device transmits a signal to one or more co-processors that the data is available to be read, wherein each of the plurality of hardware forwarding corresponds to one of the one or more co-processors. Each of these co-processors reads the data in the broadcast log by receiving the signal that the data is ready to be read from the broadcast log. In addition, each co-processor determines a broadcast log entry for the data for that co-processor. Each co-processor further reads the data from the broadcast log entry via a direct memory access in memory that stores the broadcast log and the plurality of hardware forwarding engines use the data to process network traffic.


A method for active network fabric management. The method includes receiving a probe packet by a termination beacon, where the probe packet is associated with a stream, the stream is identified using an origin beacon identification (ID) for an origin beacon, a stream source IP address, a stream destination IP address, an L2 origin interface, and a TTL value or an IP Hop value. The method further includes generating, after receiving the probe packet and after the expiration of a probe rate request (PRR) refresh timer, a rate control packet (RCP) by the termination beacon where the RCP includes a PRR for the stream, and sending the RCP to the origin beacon using an origin beacon IP address, where the origin beacon IP address is different than the stream source IP address.


Patent
Arista Networks | Date: 2016-02-09

A method and apparatus of a network element that includes a line card without retimers between an ASIC and either a network connector or mid-plane connector is described. In an exemplary embodiment, the network element includes a line card coupled to a fabric card. The line card includes a plurality of mid-place connectors, a plurality of network connectors, and a plurality of application-specific integrated circuits (ASICs). In addition, one of plurality the mid-plane connectors couple the line card with the fabric card. Furthermore, the plurality of network connectors to communicate data with devices coupled to the network element and each of the plurality of ASICs process the data. The line card further is configured such that each of the of plurality of ASICs is coupled to each of the plurality of mid-plane connectors by a different mid-plane connectorASIC path without boosting a signal on that path and each of the plurality of ASICs is further coupled to each of the plurality of network connectors by a different network connectorASIC path without boosting a signal on that path.

Loading Arista Networks collaborators
Loading Arista Networks collaborators