Glossary
Glossary of Arista NetVisor UNUM and Arista NetVisor OS Terms
Term |
Meaning |
802.1X |
The IEEE 802.1X standard provides an authentication mechanism for devices to be attached to a LAN. The authentication process involves three parties: a supplicant, an authenticator, and an authentication server. The supplicant can be a host, but it can also be a network device to be authenticated before becoming operational from a management standpoint. |
ACL |
An Access Control List is a list of rules that are used to filter network traffic and apply certain actions to it. |
AD |
Active Directory (AD). Microsoft's directory service database for Windows networks. Stores information about resources on the network and provides a means of centrally organizing, managing, and controlling access to the resources. |
Adaptive Cloud Fabric |
A group of NetVisor OS-powered switches that operate and are managed as a single holistic entity is referred to as Adaptive Cloud Fabric (Fabric in short). Later renamed to Unified Cloud Fabric. |
Alerts |
In NetVisor UNUM, alerts are special notifications directed to the user about monitored critical events. |
AOC |
An Active Optical Cable is an assembly with two transceivers that have a permanently embedded fiber optic cable. AOCs come in different fixed lengths. |
API |
An Application Programming Interface is a method to interact with NetVisor OS switches (typically in a programmatic way) that is functionally equivalent to and has a similar scope as the CLI. |
Archiver |
The NetVisor UNUM configuration Archiver module provides a convenient method of archiving NetVisor UNUM analytics and configuration data using an NFS server (for instance, for disaster recovery). It can also be used for data replication to a separate NetVisor UNUM instance to view the data offline. |
ARP |
The Address Resolution Protocol is an IETF standard protocol used to discover the link layer address, such as a MAC address, associated with a given IPv4 address. |
AS |
An Autonomous System is a collection of connected IP routing prefixes under the control of one or more network operators on behalf of a single administrative entity or domain. |
Audit Log |
An audit log is a list of events used for auditing/reviewing purposes (for example, to review logins and logouts, or other user or switch events in NetVisor UNUM instances). |
Auto-VXLAN |
A special parameter used in the VXLAN configuration to automatically assign VLAN/VNI pairs to all the VXLAN connections. Optionally, if a VNI value is not specified by the user, it is picked automatically by the system out of a predefined range. |
Bare Metal |
An industry term denoting a hardware platform without an installed operating system and in Arista Networks' case, a switch without NetVisor OS installed. |
BFD |
Bidirectional Forwarding Detection is a UDP-based protocol that provides fast detection of Layer 3 peer failures. It is used in conjunction with routing protocols to accelerate convergence in IP networks. |
BGP |
The Border Gateway Protocol is a popular standard routing protocol used to exchange routing and reach-ability data among autonomous systems. |
BMC |
The Baseboard Management Controller is a specialized micro controller in the Intelligent Platform Management Interface (IPMI) architecture. The BMC manages the interface between system-management software and platform hardware. |
Border Gateway |
EVPN Border Gateways are nodes specifically designated to run the Border Gateway functions described in detail in the Configuring EVPN chapter. Border Gateways communicate through BGP with the Border Gateways of the other pods. They also interact with the nodes within their pod. In particular, they act as translators between the intra-pod and the inter-pod communication. |
BPDU |
A Bridge Protocol Data Unit is a frame that carries information about the Spanning Tree Protocol (STP). |
Bridge Domain |
A bridge domain (BD) is an Ethernet broadcast domain in a network configuration where bridging is performed. A VLAN is a bridge domain, but the term has been extended to cover also bridging setups where the bridge domain is associated with a different identifier (such as a VNI in a VXLAN configuration) or pairs of identifiers (as in the case of VLAN stacking). |
Cgroups |
cgroups (abbreviated from control groups) is a Linux kernel feature that limits, accounts for and isolates the resource usage (such as CPU, memory, disk I/O, network) of a collection of processes. |
CLI |
The Command Line Interface is a method to interact with NetVisor OS switches through user-entered commands, which can be executed on an individual switch, on a cluster, or on a fabric. |
Cluster |
A pair of adjacent NetVisor OS-powered switches acting as one logical unit at Layer 2 for high availability. |
Cluster over Layer 3 |
This functionality supports the formation of switch cluster pairs without requiring any direct back-to-back Layer 2 connections. It leverages the uplinks connected to the spine nodes in order to form a cluster pair communicating over a Layer 3 network. |
Connections |
TCP connections are sessions established between a source and a destination to ensure that data is reliably delivered in the correct order to the upper communication layers. NetVisor OS supports collecting statistical data for TCP connections. |
CoS |
Class of Service: a 3-bit Ethernet field defined by the IEEE 802.1P standard. It’s used to define and apply eight possible levels of QoS to the traffic. |
CPTP |
Control Plane Traffic Protection is a feature used to control the traffic that reaches the CPU of a network device and to implement appropriate protections against any potentially disruptive traffic. |
CSR |
A Certificate Signing Request (or certification request) is a message sent from an applicant to a registration authority of the public key infrastructure (PKI) in order to apply for and obtain a digital identity certificate. |
DAC |
A (passive copper) Direct Attach Cable is a twin-ax (twin axial) cable assembly that connects directly into a pluggable module housing (such as an SFP+ or a QSFP+ housing). DACs are available in different fixed lengths. |
DCI |
Data Center Interconnect is a category of technologies, including those leveraging the VXLAN packet encapsulation, meant to enable the remote interconnection of data centers for improved scalability, performance and reliability or fault tolerance. |
DFM |
Dynamic Flow Mapping is a NetVisor UNUM dashboard that illustrates the total connections based on server, state, and endpoints. |
DHCP |
The Dynamic Host Configuration Protocol is a network management protocol used in IP networks to dynamically assign an IP address and other network configuration parameters to each DHCP client device (host or network node) from one or more DHCP servers. |
Distributed Multicast Forwarding with Fabric VRFs |
This feature extends the Unicast Fabric VRF functionality to route multicast traffic without employing a multicast routing protocol (such as PIM) in the underlay or overlay networks. It is also called Multicast Fabric VRFs. |
DSCP |
The Differentiated Services Code Point is a 6-bit value in the 8-bit Differentiated Services (DS) field in the IP header. It’s used for packet classification purposes for QoS and other applications. |
DNS |
Domain Name System – a hierarchical and decentralized naming system for computers, services, or other resources connected to the Internet or a private network. |
DPU |
A Data Processing Unit is an advanced system (chip or chipset) used by a high-performance network interface controller (NIC, also known as a network interface card or network adapter) that includes a programmable specialized electronic circuit with hardware acceleration of data processing for data-centric computing. See also: https://blogs.nvidia.com/blog/2020/05/20/whats-a-dpu-data-processing-unit/ |
DPU-based SmartNIC |
A DPU-based network interface card offloads processing tasks that the system CPU would normally handle. Using its own onboard processor, the DPU-based SmartNIC may be able to perform a combination of advanced networking and processing functions. See also: https://developer.nvidia.com/blog/choosing-the-best-dpu-based-smartnic/ |
ECMP |
Equal-Cost Multi-Path is a routing strategy in which next-hop packet forwarding to a single destination can occur over multiple best paths. |
ECPF |
The Embedded CPU Physical Function is also known as DPU mode and is the default mode for the BlueField DPU. In ECPF mode, the embedded Arm CPU subsystem controls the NIC resources and data path. A network function is still exposed to the host, but it has limited privileges. |
EFT |
Early Field Trial features are intended for test environments and are not supported for use in production networks. You should consult your local partner or Arista Networks account team before using any EFT feature or to provide feedback. |
eMMC |
The Embedded Multi-Media Card (MMC) architecture puts the MMC components (flash memory plus controller) into a small ball grid array (BGA) integrated circuit package for use on cards as an embedded non-volatile memory system. |
EULA |
End User License Agreement (or software license agreement) is the contract between the licensors and purchaser, establishing the purchaser's right to use the software. |
EVPN |
Ethernet VPN is a standards-based technology that leverages Multi-protocol BGP and VXLAN encapsulation to provide multi-point connectivity among different bridged domains. |
Fabric |
See Unified Cloud Fabric. |
Fabric Health |
The NetVisor UNUM platform continuously monitors networked elements across the fabric to collect extensive link layer and device-level data providing an accurate real-time picture of network and device health. Fabric Health is a dashboard that displays such information in graphical and textual format. |
Fabric Transaction |
Transactions are atomic operations that must either succeed and persist or fail and rollback, across the entire fabric. Transactions cannot be partially completed. The Unified Cloud Fabric uses transactions to synchronize configuration changes across the nodes of the fabric. |
FEC |
Forward Error Correction is a technique to detect and correct a limited number of errors in the transmitted data without the need for re-transmission. |
FHHL |
Full-Height Half-Length is a size of PCI card. |
FIB |
The Forwarding Information Base is the (software or hardware) IP forwarding table used by a switch or router to forward IP packets to their destinations. |
Filters |
Filters are a NetVisor UNUM functionality that can be applied to Reports to customize the information further. They can be created either manually or by using the spreadsheet import function. |
Firewall |
A firewall is a network security system that monitors and controls incoming and outgoing network traffic based on predetermined security policies. |
FRG |
A Fabric Resource Group is a NetVisor UNUM construct allowing the administration of groups of resources such as ports, switches, and fabrics as one logical entity. Resource Groups can be assigned to different User Groups (UG). |
FRR |
An IP routing protocol suite for Linux and Unix platforms and include protocol daemons for BGP, IS-IS, LDP, OSPF, PIM, and RIP. FR Routing (FRR) is used for connecting hosts, virtual machines, and containers to the network for network switching and routing, advertising network services, and internet peering. |
HHHL |
Half-Height Half-Length is a size of PCI card. |
Host File |
In NetVisor UNUM environments, an editable file used to preserve fabric setup information for future use. Typically host files are used in test and demonstration environments to rebuild fabrics. |
In-band Interface |
An internal interface facing the NetVisor OS kernel used as a fabric-control port when building a fabric over any IP network. |
In-band IP Address |
The IP address of the switch on a production or management network for administration and inter-switch communication. |
ICMP |
The Internet Control Message Protocol is a supporting protocol in the Internet Protocol (IP) suite. It is used by network devices, including routers, to send error messages and operational information. With IP version 6, ICMPv6 expanded its capabilities to support additional functions such as Neighbor Discovery Protocol (NDP) and Multicast Listener Discovery (MLD). |
IDS |
An Intrusion Detection System is a device or a software application that monitors the network infrastructure and/or the end devices for malicious activity or policy violations. |
IGMP |
The Internet Group Management Protocol is a communications protocol used by hosts and adjacent routers on IPv4 networks to establish multicast group memberships. |
IGMP Snooping |
IGMP snooping is a feature that enables a switch control plane to listen to IGMP messages to track group membership and to control the delivery of IPv4 multicast packets. |
Insight Analytics |
Insight Analytics is a Network Performance Management (NPM) add-on module to NetVisor UNUM. |
IPS |
An Intrusion Prevention System, also known as intrusion detection and prevention system (IDPS), is a network security appliance that monitors the network and/or the end devices for malicious activity. The main functions of an IPS are: to identify malicious activity, to log information about this activity, to report it and also to attempt to block it. |
Jumbo Frames |
Jumbo frames, or jumbos in short, are Ethernet frames with more than 1500 bytes of payload. |
Kubernetes |
Kubernetes is an open-source orchestration system for automating deployment, scaling, and management of containerized applications. |
Kubernetes Cluster |
A Kubernetes cluster consists of a control plane node and a set of worker nodes that run containerized applications. |
Kubernetes DaemonSet |
A DaemonSet manages groups of replicated Pods and ensures that all (or some) nodes run a copy of the replicated Pod. DaemonSets adds Pods to nodes that are added to a cluster and removes the Pods from nodes which are removed from the cluster. Typical use cases are: cluster storage daemons, logs collection daemons, node monitoring daemons. |
Kubernetes Deployment |
Deployments represent a set of multiple, identical Pods with no unique identities. A deployment runs multiple replicas of your application and automatically replaces any instances that fail or become unresponsive. |
Kubernetes Endpoint |
Endpoints are the set of Pods targeted by a Service. |
Kubernetes Ingress |
Ingress is API object that manages the external access to services within a cluster. Ingress exposes HTTP and HTTPS routes from outside of the cluster to services inside the cluster. An Ingress can give Services externally reachable URLs, load balance traffic, terminate SSL/TLS, and offer name-based virtual hosting. |
Kubernetes Node |
Kubernetes runs a workload by placing containers into Pods to run on Nodes. A node may be a virtual or physical machine, depending on the cluster. Each node is managed by the control plane and contains the services necessary to run the Pods. Typically, you have several nodes in a cluster; in a learning or resource-limited environment, you might have only one node. |
Kubernetes Pod |
Pods are the smallest units that you can deploy using Kubernetes. A Pod models an application-specific "logical host": it contains one or more application containers which are relatively tightly coupled (i.e., are always scheduled together). Each Kubernetes Pod has a unique IP address. |
Kubernetes Service |
A Service is an abstraction which defines a logical set of Pods and a policy by which to access them (sometimes this pattern is called a micro-service). Unlike Pods, which are non-permanent resources that can be created and destroyed dynamically, Services remain consistent. |
LACP |
Link Aggregation Control Protocol (LACP) is a protocol for the collective handling of multiple physical ports that can be seen as a single logical transmission channel (also called trunk, port channel, link aggregation group or link bundle) for network purposes such as traffic load balancing and link redundancy. It was defined in the IEEE 802.3AD standard, later incorporated into 802.3 and later moved to IEEE 802.1AX-2008. |
LAG |
Link aggregation is a technology used to combine multiple connections in order to increase the aggregate bandwidth beyond what a single connection can sustain, and to provide redundancy in case of link failure. A Link Aggregation Group (LAG) bundles a number of physical ports together to create a single high-bandwidth data path, so as to implement traffic load sharing and link redundancy. Other terms used to describe this technology include port trunking, port channel, link bundling, channel bonding "and–with" servers–NIC bonding and NIC teaming. The link aggregation process is supported by a dynamic protocol called Link Aggregation Control Protocol (LACP). |
LDAP |
The Lightweight Directory Access Protocol (LDAP) is an open, vendor-neutral, industry-standard application protocol for accessing and maintaining distributed directory information services over an Internet Protocol (IP) network. |
Layer 1 |
Also known as the Physical Layer in the OSI model, it deals with the physical transmission of data over a medium, such as Ethernet. In the Layer 1 category, for example, fall the optical transceivers used by high speed Ethernet ports. |
Layer 2 |
Also known as the Data Link Layer in the OSI model, it deals with the framing of data transmitted over a wire. Frames are logically structured sequences for data. This layer comprises two sub-layers: medium access control (MAC) and logical link control (LLC). |
Layer 3 |
Also known as the Network Layer in the OSI model, it deals with end-to-end data transmission through one or more networks. This layer implements host addressing, error checking, routing and traffic control. Network devices that support Layer 3 forwarding are called routers or Layer 3 switches. |
Layer 4-7 |
OSI model layers 4 through 7 are called: Transport, Session, Presentation and Application, respectively. They deal with the transmission of data segments between applications, the management of sessions, the data translation (when required), and the application level functions. |
LLDP |
The Link Layer Discovery Protocol is a standard link layer protocol (IEEE 802.1AB) used by network devices to advertise their identity, capabilities and neighbors on an IEEE 802 local area network. |
Log |
A log is a list of events (i.e., log entries) and related information, grouped into specific categories such as system events, port-related events, TCP traffic events, Spanning Tree Protocol events, etc. |
Loopback Trunk |
This special trunk is auto-created and is reserved to be used to carve out the necessary amount of bandwidth required for the recirculation of packets. It is used to set aside a number of (internal or front-panel) ports. |
MAC Address |
The medium access control (MAC) is a sub-layer of the Data Link layer in the OSI model as defined in the IEEE 802 LAN/MAN standards. The addresses it uses are referred to as (source and destination) MAC addresses. |
MAC Address Limit |
MAC Address Limit is a feature that restricts the maximum number of MAC addresses that can be learned on a switch port. |
Mellanox |
Mellanox Technologies Ltd. was a multinational supplier of adapters, switches, software and silicon for markets including high-performance computing, data centers, cloud computing, computer data storage. The company was integrated into NVIDIA's networking division in 2020 and NVIDIA stopped using the brand name "Mellanox" for its new networking products. |
Mesh Ping |
In NetVisor UNUM, the Mesh Ping feature validates the connectivity among all IP addresses in a fabric by executing ping tests. When complete, the dashboard quickly displays identified issues regarding connections with high latency, packet loss, or down status. |
MIB |
A Management Information Base is a database used for managing the entities in a computer network. MIBs are typically used with Simple Network Management Protocol (SNMP). |
Micro-segmentation |
NetVisor OS’s micro-segmentation supports distributed granular security policies across the fabric on DPU-based server virtual ports to bring firewall-like security capabilities all the way to the network edge. |
MLD |
Multicast Listener Discovery is a process used by IPv6 routers to discover multicast listeners on a directly attached link, much like the Internet Group Management Protocol (IGMP) is used in IPv4. |
MLD Snooping |
MLD snooping is a feature that enables a switch control plane to listen to MLD messages to track group membership and to control the delivery of IPv6 multicast packets. |
MSTP |
The Multiple Spanning Tree Protocol is a protocol introduced by the IEEE 802.1s standard and later incorporated into IEEE 802.1Q-2005, to extend the Rapid Spanning Tree Protocol to support multiple STP instances for load balancing and to introduce various other protocol enhancements. |
MTU |
The Maximum Transmission Unit is the size of the largest protocol data unit (PDU) that can be transmitted in a single network layer or data link layer transaction. |
NDP or ND |
Neighbor Discovery (Protocol) is an IPv6 node discovery process that has similar (and improved) functionalities compared to IPv4’s ARP. It is based on the ICMPv6 standard protocol. |
NetFlow |
NetFlow is a feature of Cisco routers and switches that provides the ability to collect IP network traffic statistics and to export them to a collector device. |
NetVisor OS |
NetVisor OS is Arista Networks’ enterprise-class Network Operating System (NOS) built for Open Networking hardware. NetVisor OS supports an extensive range of networking services: from the more basic ones, such as Layer 2 and Layer 3 switching for IPv4 and IPv6 protocols, to the more advanced ones such as data center interconnect (DCI) through VXLAN support and in-depth traffic analytics. |
Network Packet Broker |
A Network Packet Broker is a functionality that enables users to deploy independent traffic monitoring fabrics with a distributed architecture that supports the sharing of analytics and security tools located anywhere in the network. |
Notification |
In NetVisor UNUM a notification is a visual feedback provided in the top navigation bar with a notification bell icon about license status, fabric changes, and NetVisor UNUM errors. |
NFS |
Network File System is a distributed file system protocol that enables a user on a client computer to access files over a computer network much like local storage is accessed. |
NIC |
A Network Interface Controller (also known as a network interface card, network adapter, LAN adapter) is a computer hardware component that connects a computer (a host) to a computer network. |
ONIE |
Open Network Install Environment is an open source initiative that defines an open “install environment” acting as an enhanced boot loader for bare metal network switches. This environment allows users to install a target network OS as part of data center provisioning, in the same fashion as when servers are provisioned. |
OPN |
Ordering Part Number. |
OSPF |
Open Shortest Path First is a standard routing protocol that falls into the category of interior gateway protocols (IGPs), operating within a single autonomous system. |
OVSDB |
The Open vSwitch Database Management Protocol is an SDN configuration protocol. It is used, for example, to interface with a SDN controller such as OpenDayLight or VMware NSX. |
Out-of-band Interface |
A dedicated out-of-band port on NetVisor OS switches, used either as a management-only interface or as a fabric-control port to form the fabric and exchange fabric information over the out-of-band management network. |
Overlay |
In the VXLAN context, this term refers to all the elements built on top of the generic IP transport infrastructure in order to offer higher-level transport functionalities and services. |
PBR |
Policy-based routing is a hardware feature that is used to perform routing decisions based on policies set by the network administrator. |
PCI |
Peripheral Component Interconnect is a standard parallel local bus for attaching hardware devices in a computer. It supports expansion cards of various form factors. |
PCI Express |
PCI Express is abbreviated as PCIe or PCI-e. It’s a serial bus evolution of the PCI standard, maintained and developed by the PCI-SIG (PCI Special Interest Group). It includes six versions (as of 2022). PCIe 3.x, 4.x and 5.x are currently popular. They are also called Gen 3.x, Gen 4.x and Gen 5.x. |
PIM |
Protocol-Independent Multicast is a family of standard multicast routing protocols for IP networks that enable one-to-many and many-to-many forwarding of data over a LAN, WAN or the Internet. |
PNC |
The Pluribus Network Cloud is a portal/repository for Arista Networks software. PNC can be accessed from both the customer and partner portals under the downloads tab. |
Pod |
In a modular data center design a pod is a unit (e.g., a block of racks) that can be replicated locally and/or remotely for scalability and fault tolerance purposes. |
Port Association |
A port association is a NetVisor OS construct used to associate the status of a primary port with that of one or multiple dependent ports. In addition, it is used to create a VirtualWire connection that behaves like a pseudo-wire. |
Port Mirroring |
Port mirroring is a feature used to gain visibility into the traffic flowing through the network. It copies packets from specific source ports on a switch to destination ports on the same switch or on a different switch. Destination ports can be connected to packet sniffers or other traffic analysis tools. |
Port Storm Control |
Port Storm Control (sometimes called traffic storm control, or simply storm control) limits the percentage of the total available port bandwidth that can be used by broadcast, multicast, or flooded unicast traffic. It can be enabled to prevent excessive flooded traffic from degrading network performance. |
Pre-emphasis |
In high-speed Ethernet pre-emphasis is used to correct the distortion introduced by the transmission medium. The result is to improve signal quality at the output of the data transmission. |
PXEBOOT |
The Preboot eXecution Environment, PXE (pronounced as “pixie”), is often called PXE Boot. It’s a specification that describes a standardized client–server environment that boots a software retrieved from a network on PXE-enabled clients. |
Quagga |
Quagga is a network routing software suite providing implementations of Open Shortest Path First (OSPF), Routing Information Protocol (RIP), Border Gateway Protocol (BGP) and IS-IS for Unix-like platforms. |
QinQ |
QinQ is a technique (also known as stacked VLANs, or Q-in-Q) that can apply an extra VLAN tag on top of the standard 802.1Q tag (hence the term of VLAN stacking). |
QoS |
Quality of Service refers to traffic prioritization and resource reservation control mechanisms that can provide different priorities to different applications, users, or data flows, and that can guarantee a certain level of performance to each data flow. |
QSFP+ |
The Quad Small Form-factor Pluggable module is a compact, hot-pluggable transceiver used for data communications applications. QSFP+ is an evolution of QSFP to support four channels carrying 10 Gigabit Ethernet that can be combined to form a single 40 Gigabit Ethernet link. |
QSPF28 |
A Quad Small Form-factor Pluggable module that supports speeds of 100 Gbit/s with encoding overhead (e.g., to carry 100 Gigabit Ethernet). |
QSPF56 |
A Quad Small Form-factor Pluggable module that supports speeds of 200 Gbit/s with encoding overhead (e.g., to carry 200 Gigabit Ethernet). It’s backward compatible with QSFP28. |
RA |
Router Advertisement is a type of ICMPv6 message used for the Neighbor Discovery (ND) process. |
Report |
In NetVisor UNUM, an (email) report is an organized list of information derived from a dashboard or from search criteria generated in various formats (such as PDF, PNG, or CSV) and sent to a receiver. |
RESTful |
Representational State Transfer (REST) is a software architectural style that defines a set of rules to be used for creating web services. Web services that conform to the REST architectural style are called RESTful. |
RIB |
The Routing Information Base is the IP routing table created by a switch or router by collecting routing information from multiple sources including configuration (static routes), dynamic routing protocols (RIP, OSPF, BGP), etc. |
RIP |
The Routing Information Protocol is an old distance-vector routing protocol that employs the hop count as a routing metric. It has two versions, RIPv1 and RIPv2, for IPv4 while RIPng is an extension of RIPv2 with support for IPv6. |
RMA |
A Return Merchandise Authorization is usually referred to the process of returning a product to receive a replacement or repair (and implicitly following the associated network administrator procedures). |
RSTP |
The Rapid Spanning Tree Protocol was introduced as standard IEEE 802.1w to provide significantly faster spanning tree convergence after a topology change compared to regular STP, while maintaining full backward compatibility with it. |
SCP |
The Secure Copy Protocol is a network protocol based on the BSD RCP protocol that supports secure file transfers between devices on a network. Security (authenticity and confidentiality of the data in transit) is based on the Secure Shell (SSH) protocol. |
Seed Switch |
In Arista Networks' software-defined fabrics, a seed switch is a fabric node that acts as a central point of communication between NetVisor UNUM and the fabric. The seed switch is where the NetVisor UNUM collector resides. |
SDN |
Software-Defined Networking is defined by the Open Networking Foundation as an emerging architecture that is dynamic, manageable, cost-effective and adaptable, making it ideal for the high-bandwidth, dynamic nature of today's applications. |
sFlow |
sFlow stands for ‘sampled flow’. It is an industry standard for traffic sampling and exporting of truncated packets, together with interface counters, for the purpose of network monitoring. |
sFlow Agent |
The sFlow Agent is a software component that runs on an Arista switch and uses sampling technology to capture traffic statistics directly from the switch's hardware. sFlow datagrams are used to immediately forward the sampled traffic statistics to a central sFlow Collector for analysis. |
sFlow Collector |
An sFlow Collector is a centralized device that receives sFlow datagrams from one or more sFlow Agents. |
sFlow Counter Polling |
Periodic polling of counters associated with a data source. NetVisor OS uses a configurable timer for counter polling: on its expiry it collects the statistics from the hardware and with them it constructs an sFlow sample that sends to a central sFlow Collector for analysis. |
sFlow Packet Flow Sampling |
Packet Flow Sampling refers to the random selection of a fraction of the packets in the flows observed at an sFlow data source. |
sFlow Sample Rate |
Sample rate (or sampling rate) is the ratio of packets observed at the data source to the samples generated. For example, a sample rate of 100 specifies that 1 sample will be generated for every 100 packets observed, on average. |
SFP |
A Small Form-factor Pluggable is a compact, hot-pluggable network interface module. It can support various speeds, but it’s most commonly used in various 1 Gbit/s flavors, either with fiber-optic cables or with copper cables. It can be multi-rate as in the 100/1000BASE-T case (100/1000 MBit/s). |
SFP+ |
The enhanced Small Form-factor Pluggable module is a compact, hot-pluggable transceiver that supports data rates of up to 16 Gbit/s and is therefore used for 10 Gigabit Ethernet interfaces. |
SFP28 |
A small form-factor pluggable (SFP) that supports speeds of 25 Gbit/s with encoding overhead (e.g., for 25 Gigabit Ethernet) and is backward compatible with SFP+. |
SFP56 |
A small form-factor pluggable (SFP) that supports speeds of 50 Gbit/s with encoding overhead (e.g., for 50 Gigabit Ethernet) and is backward compatible with SFP28. |
SFTP |
The SSH File Transfer Protocol is an IETF network protocol that provides file access, file transfer and file management over any reliable data stream. It was designed as an extension of the Secure Shell protocol (SSH) version 2.0 to provide secure file transfer capabilities. |
SLACC |
IPv6 State-Less Address Auto-Configuration is the process by which IPv6 hosts can configure themselves automatically when connected to an IPv6 network using the Neighbor Discovery Protocol function via ICMPv6 router discovery messages. |
SNMPv3 |
Simple Network Management Protocol is an IETF standard protocol for collecting and organizing information about managed devices on IP networks and for modifying that information to change a device’s behavior. SNMPv1 is the original version of the protocol, while SNMPv2c and SNMPv3 are more recent versions that feature improvements in performance, flexibility and security. |
Splitter Cable |
A splitter cable (also called breakout cable) is a copper or optical cable assembly with a QSFP on one end and 4 SFPs on the other hand. It can be a QSFP+ module connected to four SFP+ modules on the other end. Or it can be a QSFP28 module connected to four SFP28 modules. A splitter cable can be of the DAC (copper) or AOC (fiber optic) type. |
SSH |
Secure Shell is a cryptographic network protocol that enables network services to operate securely over an insecure network. Typical applications include remote command-line login and remote command execution. |
Static Routing |
Static routing refers to the configuration in NetVisor OS or in NetVisor UNUM of static route table entries, which are then used by the hardware to forward the traffic. |
STP |
In a bridged network the Spanning Tree Protocol (IEEE 802.1D and IEEE 802.1Q-2014 standards) is used to turn a redundant physical topology into a loop-free, tree-like logical forwarding topology by setting one or more ports to blocking state, so as to prevent bridging loops. |
Switch Analytics |
In NetVisor UNUM, Switch Analytics is a suite of tools that enable administrators to monitor system, traffic, and connectivity statistics, manage syslog notifications, and display active endpoints or vPorts statistics. |
Switch Mode |
NetVisor OS switches can be used as regular fabric nodes for traffic forwarding or as VirtualWire connection nodes. The two mutually exclusive switch modes are called store-and-forward and virtual-wire. |
Syslog |
Syslog is a standard technology for message logging which logically separates the software that generates the messages, the system that stores them, and the software that reports them. |
TACACS+ |
Terminal Access Controller Access-Control System Plus (TACACS+) is an open protocol that handles authentication, authorization, and accounting (AAA) services. It was originally described in an Internet Draft (draft-grant-tacacs-02.txt). |
TLS |
Transport Layer Security is a cryptographic protocol designed to provide communication security over a computer network with the aim of guaranteeing privacy and data integrity between two or more communicating computer applications. |
Traffic Flow |
Also known as packet flow or network flow, is a sequence of packets from a source device to a destination (a unicast destination, a multicast group, or a broadcast address). |
Topology |
The term network topology refers to the network structure, including the arrangement of the network nodes and their interlinking. In NetVisor UNUM, the topology dashboard provides a graphical representation of such a structure. |
Topology Configuration |
A topology configuration is a feature can be used to automate the creation of vLEs or vPGs in a repeatable, convenient and user-friendly fashion (e.g., without having to save the entire device’s configuration and without having to restart a device when the configuration changes). |
Transceiver |
An Ethernet transceiver is a hardware component (in a fixed or modular form factor) that implements transmission and reception at the physical layer (Layer 1). In other words, a transceiver is both a transmitter and a receiver, hence the name. |
Trunk |
A Link Aggregation Group. Sometimes a trunk is also called a (port) trunk, link bundle, or port channel as it bundles a number of connected physical ports together to implement a single augmented logical communication channel capable of traffic load sharing with fast re-convergence and redundancy. |
UG |
A User Group manages access to NetVisor UNUM dashboards and menus. It contains users, admins, and assigned privileges. |
Unified Cloud Fabric |
A group of NetVisor OS-powered switches that operate and are managed as a single holistic entity is referred to as Unified Cloud Fabric (Fabric in short). Unified Cloud Fabric was previously known as Adaptive Cloud Fabric. |
Uplink Group |
The uplink group configuration allows users to track the aggregate bandwidth available to a group of physical uplink ports. It also allows the users to select which action to apply in case of aggregate bandwidth mismatch between the uplink ports and the downlink ports. |
Underlay |
In the VXLAN context, this term refers to the generic IP transport infrastructure used to ensure IP reach-ability among all Virtual Tunnel Endpoints (VTEP) in the network that create the overlay. |
Unicast Fabric VRF with Anycast Gateway |
NetVisor OS supports the VRF technology in hardware allowing multiple routing spaces to coexist on the same distributed fabric architecture. With the addition of the Anycast Gateway functionality, the Unified Cloud Fabric enables distributed forwarding at the first hop router as well as intrinsic VM mobility capabilities across complex multi-site data center designs. |
NetVisor UNUM |
NetVisor UNUM Unified Management, Automation and Analytics Platform is a multi-functional web management portal that enhances the intrinsic automation of the Unified Cloud Fabric architecture. |
vCenter Server |
vCenter Server is the centralized management utility for VMware. It is used to centrally manage hypervisors (ESXi), storage, virtual machines, and all dependent components (such as network and security). |
vFlow |
Arista Networks’ mechanism used to filter fabric-wide data center switching traffic on a granular flow level, and to apply security/QoS (Quality of Service) actions or forwarding decisions on each defined flow. |
VIP |
A Virtual IP is an IP address that does not correspond to an actual physical device but to a virtual forwarding entity (for example for redundancy purposes). In this document’s context it’s the IP address used by VRRP instances and by VTEPs. |
VirtualWire Connection |
A VirtualWire connection is a pseudo-wire feature that emulates a physical wired connection so that two or more physical ports in a single switch or across a multi-switch topology get logically interconnected in a point-to-point or point-to-multi-point fashion. This pseudo-wire feature does not require moving the cables around and is a powerful capability in a network lab to remotely control device interconnections. A VirtualWire connection can be of three types: a L1 mode connection, a vLE or a vPG. |
VLAG |
Virtual Link Aggregation Group is a NetVisor OS multi-chassis link aggregation technology to bundle two or more links together when the links belong to two different chassis (behaving as a single virtual chassis/cluster). |
VLAN |
A Virtual LAN is a logical broadcast domain that is identified by using a specific frame tag format (defined by the IEEE 802.1Q standard) and is isolated at the data link layer in a computer network. |
vLE |
Virtual Link Extension is a NetVisor OS technology that enables the creation of pseudo-wires that can emulate a direct connection between devices on top of an IP transport network. A vLE is one kind of VirtualWire connection (see its definition above). |
VMWare Dashboards |
NetVisor UNUM supports Insight Analytics for VMware virtual machine traffic and connection statistics (for ESXi servers, vCenter instances, etc.). In addition, it supports a Dynamic Flow Mapping dashboard to illustrate the total connections based on server, state, and endpoints. Currently, VMWare dashboards are an Early Field Trial feature. |
vNET |
A Virtual NETwork is a partition of the fabric. A vNET is defined by a group of network objects that can operate independently and have dedicated resources. This is how NetVisor OS provides multi-tenancy support and in-depth network segmentation (beyond VLANs and VRFs). |
VNI |
In VXLAN parlance, each segment is identified through a 24-bit segment ID called the “VXLAN Network Identifier” (VNI). This allows up to 16M VXLAN segments to coexist within the same administrative domain. |
vPG |
A Virtual Port Group construct enables users to transport the traffic that arrives at one or more select source ports to one or more desired destination ports, unidirectionally or bidirectionally. A vPG can be used to create a VirtualWire connection or in a Network Packet Broker deployment. |
vPorts |
“Virtual ports” are software Layer 2 entries associated to all ports an Arista switch performs MAC address learning on. |
VRF |
Virtual Routing and Forwarding is a technology that allows multiple routing spaces to coexist on the same switch. It complements the vRouter construct, offering a highly scalable solution for multi-tenant environments. (See also Fabric VRFs) |
vRouter |
An object used to provide routing between subnets, VLANs and/or vNETs. The vRouter runs in a dedicated operating system container. |
vRouter Interface |
There are two types of interfaces that can be created on a vRouter object to route traffic: physical and virtual (Layer 3) interfaces. The former type corresponds to physical ports used to connect to other devices, such as other Layer 3 switches or routers. The latter type corresponds to VLAN interfaces (a.k.a. SVIs) used to enable routing for VLANs/subnets. |
VRRP |
Virtual Router Redundancy Protocol is a networking protocol that provides redundancy of routing paths by creation of virtual routers, which are an abstract representation of multiple routers (i.e., master and backup routers) acting as a group. |
vSG |
A Virtual Service Group (vSG) is a feature that enables a user to share subnet prefixes between Unicast Fabric VRFs in a simple and user-friendly way. |
VTEP |
A VXLAN Tunnel Endpoint is the entity responsible for encapsulating / de-encapsulating VXLAN packets. |
VTEP HA |
VTEP High Availability refers to a mechanism designed to ensure redundancy of the VTEP entity. |
VXLAN |
Virtual Extensible LAN is a standard UDP-based packet encapsulation technology defined in RFC 7348. VXLAN’s Ethernet-in-UDP encapsulation is used to implement the overlaying of virtualized Layer 2 networks over Layer 3 networks. |
WEB-HA |
Arista Networks' Unified Cloud Fabric provides High Availability (HA) for REST API services (“web services”) through VRRP running on a switch cluster pair. Redundant seed switches operate in an active / standby manner accessed via a common VRRP VIP (Virtual IP address). |
Wireshark |
Wireshark is a free open source packet analyzer. It is used for network troubleshooting and analysis, and for software and communications protocol development. |
ZTP |
Zero Touch Provisioning is a network device capability that enables it to be provisioned and configured automatically, reducing the overhead required for a complete network deployment. |