NS0-304 key terms

A set of up to many nodes that present a single administrative and namespace view, share configuration through the cluster database, and can fail workloads over to partner nodes. Clients connect to logical interfaces rather than physical node ports.

A logically isolated server inside an ONTAP cluster that owns its own volumes, LIFs, namespaces, and access policies. Each SVM serves data independently, letting one cluster host many separate tenants or workloads.

The replicated configuration store that every node holds a copy of and that ONTAP uses to keep cluster-wide settings consistent. Changes are committed cluster-wide, so nodes share one authoritative view of configuration.

A voting mechanism that keeps a cluster operational only while a majority of nodes can communicate. Epsilon is the extra tie-breaking vote assigned to one node so that even-sized clusters can still reach a majority.

A network address with a role, such as data, cluster management, node management, or intercluster, that can move between ports and nodes. Clients use data LIFs, so failover and migration stay invisible to them.

A separate address and routing domain inside a cluster, with its own IP addresses, subnets, and broadcast domains. IPspaces let overlapping address ranges serve different tenants without conflict.

A grouping of cluster ports that share Layer 2 reachability, used to constrain where a LIF may be placed or failed over. Ports in the same broadcast domain can host the same LIF.

A named set of ports and nodes that a data LIF is allowed to move to during a takeover or planned migration. It bounds where the address can appear so clients keep reachability.

A collection of physical disks or disk partitions that ONTAP manages as one storage pool from which volumes are carved. Aggregates define RAID protection and provide the capacity and performance backing for flexible volumes.

A double-parity RAID layout that survives two simultaneous disk failures in the same RAID group. It is the common default protection for ONTAP aggregates and trades some capacity for resilience.

A triple-parity RAID layout that tolerates three simultaneous drive failures, typically used with very large drives where rebuild times are long. It improves protection at the cost of additional parity capacity.

A disk not assigned to any aggregate that ONTAP can use to rebuild a failed drive automatically. Keeping adequate spares of the right type and size is required for reconstruction and for growing aggregates.

Background checks that read disk data to detect and repair media errors before they affect a rebuild. Scrub validates parity and RAID consistency, while media scan looks for bad blocks on the drives.

A flexible volume contained in an aggregate that can grow or shrink within available space and shares the aggregate's performance. It holds files in a namespace or blocks exported as LUNs.

A single namespace volume made of many constituent member volumes spread across nodes, giving high capacity and parallel throughput. It scales performance and size while presenting one mount point to clients.

The mount point inside an SVM namespace where a volume appears, such as /data/projects. Junctions build a hierarchical namespace so NFS and SMB clients see volumes as directories.

A space-efficient, read-only point-in-time image of a volume that preserves the state of its blocks without copying all data. Snapshots underpin backups, restores, and replication.

A named schedule that determines when ONTAP creates snapshots and how many are retained on a volume. Applying a policy automates protection and controls snapshot space consumption.

A percentage of volume space set aside for snapshot copies so they do not consume space intended for active data. When the reserve fills, older snapshots are typically deleted according to policy.

ONTAP replication that copies snapshots from a source volume or SVM to a destination for disaster recovery, either asynchronously or synchronously. The destination is normally read-only until a failover makes it writable.

A replication relationship that transfers snapshots from a primary to a secondary system for longer retention and backup, rather than for immediate failover. It is commonly used for disk-to-disk backup copies.

A rule set that defines which snapshots are replicated and how, such as asynchronous mirroring or vaulting, plus retention behavior on the destination. It determines replication scope and recovery options.

SVM-level replication that copies configuration and volumes together so a whole SVM can be served from the destination after failover. It reduces recovery effort compared with replicating volumes individually.

A block storage object inside a volume that a SAN host sees as a disk. ONTAP serves LUNs over iSCSI, FC, or NVMe and controls access through igroups and LUN maps.

A named collection of host initiators, identified by iSCSI IQN or FC WWPN, used to grant SAN access. LUNs are mapped to igroups rather than to individual hosts.

The association between a LUN and an igroup that permits the listed initiators to see and use the LUN. Correct mapping and matching host operating system types prevent access and alignment problems.

Asymmetric logical unit access lets a SAN host prefer the node that owns a LUN and use the partner path only when needed. ONTAP reports path states so multipathing software can optimize traffic.

A set of rules that decides which NFS clients may mount an SVM volume and with what access, based on client address and export options. It is the NFS counterpart to SMB share permissions.

A named resource published by an SVM that Windows and other SMB clients connect to, governed by share permissions and NTFS file permissions. It maps to a path inside the SVM namespace.

The SMB server identity that an SVM presents, including its NetBIOS and DNS names and its authentication domain membership. Clients authenticate to this server when accessing SMB shares.

The ordered list that tells an SVM which sources to consult for name lookups, such as local files, DNS, LDAP, or NIS. It controls how users, groups, and hosts are resolved for access decisions.

ONTAP's authorization model that grants users or groups specific commands and API access through predefined or custom roles, scoped to a cluster or SVM. It limits administrative actions to what a role permits.

Requiring more than one proof of identity, such as a password plus a time-based code, before granting administrative access. ONTAP supports it to reduce risk from stolen credentials.

Recording file access and administrative events so activity can be reviewed for compliance and investigations. ONTAP can write audit records to configured destinations and rotate them by policy.

The process where a node's partner assumes its storage and LIFs during a failure, then returns them when the node recovers. It provides high availability for data access in a cluster.

A nondisruptive operation that relocates a volume to a different aggregate, optionally on another node, while clients keep accessing it. It is used for rebalancing capacity and performance.

AutoSupport sends diagnostic and health data from the system to support, and Active IQ analyzes it to surface risks, best practices, and predictive alerts. Together they support proactive monitoring.

A set of throughput limits or shares applied to volumes or workloads so that one workload cannot starve others. It controls maximum throughput and helps meet performance objectives.

Automated nondisruptive upgrade software that updates cluster nodes in sequence, moving workloads to partners so clients keep working. It validates the cluster against upgrade requirements before proceeding.

The two main automation interfaces: a documented REST API for programmatic calls and the command-line shell for scripting and interactive work. Automation tools typically combine them with SDKs.

Libraries and modules that let scripts and orchestration tools create volumes, shares, and protection relationships without manual steps. They call ONTAP APIs and support repeatable, version-controlled provisioning.

A software-only ONTAP deployment running in public clouds that provides the same data management features as on-premises systems. It uses cloud storage and integrates with cloud networking and identity services.