NCSE-Core flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

Controller VM (CVM)

A VM on each AHV host that owns the node's local disks and serves storage I/O to all nodes over the internal network, forming the distributed storage fabric.

Prism Element vs Prism Central

Prism Element manages a single cluster (hardware, storage, VMs, alerts); Prism Central aggregates multiple clusters and hosts Flow, NCM, Files, and DR features.

Genesis

The Nutanix bootstrap and imaging service used by Foundation to image nodes and perform initial configuration before AOS cluster creation.

Replication Factor (RF)

The number of copies of data kept per storage container. RF2 survives one disk or node failure; RF3 survives two. Usable capacity is roughly raw divided by RF.

Curator

The background DSF scanner that detects disk failures, rebuilds missing replicas, balances disks, and performs garbage collection on the distributed storage fabric.

Stargate

The DSF I/O layer that writes extents locally and replicates them to remote CVMs before acknowledging writes, preserving data on node loss.

Erasure Coding (EC-X)

A capacity-saving scheme (e.g., 4+1, 6+2) that reduces overhead versus RF but increases write amplification, so it is typically applied to cold data.

AHV high availability

After a host failure, AHV restarts affected VMs on surviving hosts. It preserves disk state but not in-memory state and is not guest clustering.

Nutanix Guest Tools (NGT)

In-guest tools that enable VSS-consistent snapshots, self-service restore, and file-level restore for Windows VMs on AHV.

Affinity policies

Rules that pin VMs to hosts (VM-host affinity) or keep VMs apart/together (VM-VM anti-affinity/affinity); enforced during HA and live migration.

acli

The Acropolis CLI on the CVM used to manage AHV VMs, hosts, and images when the Prism UI or API is unavailable.

balance-slb bond mode

Default AHV uplink bond that hashes per flow across links without switch configuration; works with any switch but does not aggregate a single flow.

Jumbo frames

MTU 9000 configured end-to-end on CVM and host management networks to improve storage throughput; inconsistent MTU causes silent performance and connectivity issues.

LACP bond mode

Active-active AHV uplink bond requiring a switch LAG; provides higher aggregate throughput and link redundancy but needs matching switch configuration.

Flow Network Security

Prism Central-managed distributed firewall that enforces policy at the vNIC on AHV hosts, enabling microsegmentation by VM or category.

Microsegmentation

Applying allow-list firewall policy to individual workloads or tiers, typically scoped by categories, to limit east-west lateral movement inside the data center.

Flow Virtual Networking (VPC)

Overlay virtual networks managed from Prism Central with their own IP space, routing, NAT, and optional external connectivity, isolated per tenant or application.

Categories

Key-value metadata assigned to VMs and clusters that drives policy scoping, protection domains, playbooks, and Flow security rules without per-VM edits.

REST API v3

The intent-based Prism Central API (/vms, /clusters, /images) used for automation; Prism Element exposes the older v2 API.

NCC health checks

Nutanix Cluster Check runs diagnostic checks on cluster health, configuration drift, and upgrade readiness; output is used for support cases.

Life Cycle Manager (LCM)

Prism feature that inventories and upgrades firmware and software components, checking compatibility before applying updates.

Protection domain

A group of VMs, often defined by category, with snapshot schedules, retention, replication targets, and consistency settings for data protection.

RPO vs RTO

RPO is the maximum acceptable data loss (driven by replication interval); RTO is the maximum acceptable downtime (driven by recovery method).

Nutanix DR runbooks

Recovery plans in Prism Central that boot VM groups in order with IP and guest customization for planned, unplanned, or test failover.

Metro availability

A stretched cluster across two sites with a third-site witness for quorum, providing near-zero RPO at low-latency links (typically under 5 ms RTT).

Nutanix Move

A migration tool that moves VMs from ESXi, Hyper-V, AWS, or Azure to AHV with pre-checks, data seeding, and minimal-downtime cutover.

Nutanix Cloud Clusters (NC2)

AOS and AHV running in AWS or Azure, managed by Prism, used for cloud bursting, DR, and lift-and-shift with VPC/VNet networking.

Cutover planning

Freeze source, final delta sync, power on target, validate the application, then decommission source after a documented rollback window.

Usable capacity formula

Usable capacity is roughly raw capacity divided by replication factor, minus reserved space, snapshot overhead, and rebuild headroom of about one node.

Compute sizing ratio

Total vCPU demand divided by a target vCPU-per-core ratio (often 4:1 to 6:1) gives cores needed; memory is usually the binding constraint.

N+1 host headroom

Design capacity so that losing one host still leaves enough CPU and memory to run all workloads, matching AHV HA restart requirements.

Pre-sales discovery inputs

VM inventory, performance baselines, growth rate, RTO/RPO, compliance needs, and network topology; these drive sizing, resilience, and the bill of materials.