Cisco DCID 300-610 key terms

A 24-bit value in the VXLAN header that identifies a virtual Layer 2 segment, letting a single physical network carry up to about 16 million isolated segments instead of the roughly 4,000 available with VLAN IDs.

The device or interface that encapsulates and decapsulates VXLAN traffic, mapping frames between a local Layer 2 segment and an IP-routed overlay tunnel toward a remote VTEP.

A logical Layer 2 forwarding construct in Cisco ACI that groups ports and maps to one VXLAN VNI, controlling flooding and learning within a tenant while subnets and routing are configured separately.

An ACI policy object that groups endpoints sharing the same network policy requirements, such as a tier of application servers, so that contracts can be applied to the group rather than to individual ports.

An ACI policy construct defining which protocols and ports are permitted between endpoint groups, implemented through subjects, filters, and one or more directional provider-to-consumer relationships.

A separate routing table instance that isolates Layer 3 reachability, allowing overlapping IP subnets to coexist on shared infrastructure while keeping traffic paths distinct per tenant or function.

The pair of Cisco Nexus switches in an ACI fabric that act as the fabric's control plane, policy repository, and external connectivity point, running APIC-managed fabric services.

The cluster of controllers that manages Cisco ACI fabrics, storing the policy model, programming switches, and providing a single point for fabric-wide configuration and health monitoring.

A two-tier data center topology in which every leaf switch connects to every spine, giving predictable latency and equal-cost multipathing while keeping server-facing ports on leaves only.

A Cisco feature that lets two switches present a single logical port channel to a downstream device, providing link redundancy and active-active forwarding without spanning tree blocking.

An aggregation of multiple physical links into one logical interface using LACP or static bundling, increasing bandwidth and providing link-level redundancy between two network devices.

Two ways VXLAN fabrics discover remote endpoints: flood-and-learn replicates unknown traffic and learns from responses, while controller-based learning distributes endpoint locations through a central control plane.

The management application embedded in Cisco UCS fabric interconnects that provisions servers, pools identities, and applies service profiles across all attached chassis from a single interface.

A UCS object containing all server configuration, including identity, firmware, network, and storage settings, which is associated with hardware so a replacement server can inherit identical settings.

A collection of reusable identifiers such as MAC addresses, WWNNs, WWPNs, or UUIDs from which UCS assigns values to service profiles, enabling consistent server identity across hardware replacements.

A UCS fabric interconnect port that can carry both Ethernet and Fibre Channel traffic, negotiated by the adapter, reducing cabling by supporting LAN and SAN on the same link.

The UCS component that connects chassis and rack servers to LAN and SAN uplinks, hosts UCS Manager, and provides switching plus identity and policy enforcement for attached servers.

A cloud-based management platform for Cisco compute and infrastructure that provides policy-based configuration, firmware management, health monitoring, and automation across data center and edge systems.

A UCS enclosure holding multiple blade servers plus shared power, cooling, and I/O modules, connected to fabric interconnects and managed as a unit within a UCS domain.

A UCS construct that assigns a set of fabric interconnect Ethernet uplink ports to a vNIC template or LAN connectivity policy, controlling which uplinks carry a given server's traffic.

A reusable UCS definition of virtual network interface settings such as fabric placement, VLAN membership, MTU, and failover, applied to service profiles for consistent server networking.

A standard that encapsulates Fibre Channel frames inside Ethernet frames so storage traffic can share converged links, requiring lossless Ethernet configuration such as priority flow control.

A virtual SAN that partitions a Fibre Channel fabric into isolated logical fabrics, each with its own zoning and fabric services, allowing separate storage traffic to share physical switches.

A Fibre Channel fabric service that defines which initiators may communicate with which targets by grouping device ports into zones, enforced by the fabric during login and name server queries.

A globally unique 64-bit identifier assigned to a Fibre Channel port, used by fabric zoning and name servers to identify initiators and targets independent of physical switch location.

A Cisco feature that lets an edge switch act as an N_Port proxy toward the core, reducing domain ID consumption and switch management while still passing host logins upstream.

A named label mapped to a WWPN or Fibre Channel ID in Cisco MDS and Nexus switches, letting zoning and configuration reference friendly names instead of raw identifiers.

A protocol that carries SCSI commands over TCP/IP, allowing hosts to access block storage across standard Ethernet networks without a dedicated Fibre Channel fabric.

Host-side software that presents multiple physical paths to the same storage LUN as one logical device, providing failover and load balancing when a path or fabric component fails.

A family of protocols such as HSRP, VRRP, and GLBP that give multiple routers a shared virtual gateway address, so hosts keep connectivity when a gateway device fails.

A design where the same default gateway IP and MAC are configured on every leaf switch, so a host's traffic is routed locally wherever the host attaches, avoiding traffic tromboning.

Ethernet frames with an MTU larger than the standard 1500 bytes, commonly around 9000 bytes, used to reduce per-packet overhead for storage and bulk data transfers.

The point in the network where packet markings such as DSCP or CoS are accepted as authoritative, typically set at the server access port so downstream devices preserve classification.

An Ethernet mechanism that pauses only a specific traffic class instead of the whole link, enabling lossless behavior for storage traffic on converged links without stalling other traffic.

The practice of defining network and compute configuration in version-controlled files that automation tools apply consistently, enabling repeatable builds, review, and rollback of changes.

An agentless automation tool that applies declarative playbooks over SSH or APIs to configure network and compute devices, commonly used for repeatable data center provisioning tasks.

Continuous push of operational data such as counters and interface state from devices to a collector, giving near-real-time visibility compared with periodic SNMP polling.