Web Hosting Industry Terminology

Anycast

A network addressing and routing method where the same IP address is announced from multiple locations, and BGP routes users to the nearest or best path endpoint to reduce latency and improve resilience.

We publish our DNS resolvers via Anycast for faster lookups; Our CDN uses Anycast IPs to steer traffic to the closest PoP; Our DDoS mitigation relies on Anycast scrubbing centers

Autoscaling

Automatically adjusts compute resources (instances, containers, pods) based on load or metrics to maintain performance and control costs.

Enable autoscaling to handle traffic spikes; We configured min and max nodes in the autoscaling group; Policies scale out when CPU exceeds 60%

Backup and Disaster Recovery (DR)

Strategies and processes to copy data and restore systems after failures or disasters, including backup schedules, retention, offsite storage, and failover plans.

We run nightly backups with 30-day retention; We test DR runbooks every quarter; Our DR design uses a warm standby in another region

Bandwidth and 95th Percentile Billing

A billing method that samples throughput over time, discards the top 5% of peaks, and bills on the highest remaining sample to encourage consistent usage and reduce penalty for brief spikes.

95th percentile billing reduced our overage charges; We sample bandwidth at 5-minute intervals; Peering lowered our 95th percentile bill

BGP (Border Gateway Protocol)

The interdomain routing protocol that exchanges reachability information between autonomous systems, controlling how traffic flows on the Internet.

We announce our prefixes to two transit providers via BGP; Anycast depends on BGP route selection; We use BGP communities to influence inbound routes

CAC (Customer Acquisition Cost)

The average cost to acquire a new paying customer, including marketing and sales expenses.

Our LTV:CAC must be above 3:1; Include ads, commissions, and sales salaries in CAC; Organic SEO helped lower CAC

CDN (Content Delivery Network)

A geographically distributed network that caches and serves content from edge locations to reduce latency, offload origin servers, and improve performance.

Put static assets behind the CDN; Purge the CDN cache during deploys; Anycast CDN PoPs improved TTFB

Churn Rate

The percentage of customers or revenue lost in a given period, a key metric in subscription hosting businesses.

Monthly churn fell from 2% to 1.4%; Involuntary churn rose due to failed renewals; Better onboarding reduced early churn

Cloud Hosting

Using virtualized, on-demand infrastructure and services (compute, storage, networking) delivered by a cloud provider, billed pay-as-you-go.

We migrated from dedicated servers to cloud hosting for elasticity; Burst traffic is a good fit for cloud; We adopted a managed cloud database

Containerization

Packaging applications and their dependencies into isolated, portable units (containers) that run consistently across environments.

We package our PHP app into a container; Images are stored in a private registry; Each site runs in an isolated container

cPanel

A widely used web hosting control panel that simplifies site, email, DNS, and server management for shared and reseller hosting.

Create email accounts in cPanel; Install WordPress via cPanel’s auto-installer; Manage DNS zones through cPanel

DDoS (Distributed Denial of Service)

An attack that floods a target with massive traffic or requests to exhaust resources and disrupt availability.

We enabled upstream DDoS protection; Rate limiting on the load balancer helped; Anycast diverted volumetric attacks

Dedicated Server

A physical server allocated to a single customer, providing full resource control and isolation compared with shared or virtual hosting.

We upgraded to a dedicated box for compliance; RAID 10 protects the dedicated server’s disks; Add remote hands support for the fleet

DNS (Domain Name System)

The distributed system that translates domain names into IP addresses and other records (A, AAAA, CNAME, MX, TXT) for service discovery.

Lower the DNS TTL before a migration; Use DNSSEC to sign the zone; Create A and AAAA records for the load balancer

Email Deliverability

The likelihood that emails reach recipients’ inboxes rather than spam, affected by authentication, reputation, and content.

Configure SPF, DKIM, and DMARC; Warm up new sending IPs; Check blocklists when diagnosing spam placement

Firewall (Network Firewall)

A security control that filters network traffic using allow/deny rules based on IPs, ports, and protocols.

Only allow inbound 80/443 and SSH; Apply egress filtering to limit outbound access; Use stateful firewall groups per tier

GDPR (General Data Protection Regulation)

EU regulation governing personal data protection and privacy, with strict requirements and penalties for non-compliance.

Sign DPAs with EU customers; Enable data residency in EU regions; Honor data subject access requests

High Availability (HA)

Designing systems with redundancy and failover so services remain available despite component failures.

Active-active app nodes behind a load balancer; Multi-AZ databases for HA; Eliminate single points of failure

HTTP/2 and HTTP/3

Modern versions of HTTP that improve performance via features like multiplexing (HTTP/2) and QUIC transport (HTTP/3).

Enable HTTP/3 with QUIC on NGINX; HTTP/2 multiplexing reduced head-of-line blocking; ALPN negotiates h2 over TLS

Hypervisor

The software layer that allows multiple virtual machines to run on a single physical host by abstracting hardware resources.

KVM is our hypervisor; Manage guest VMs via the hypervisor API; Hypervisor overhead stays under 5%

IaaS/PaaS/SaaS

Cloud service models: Infrastructure as a Service (VMs, storage, networking), Platform as a Service (managed runtimes), and Software as a Service (applications).

Our VPS offering is IaaS; We added a PaaS for managed runtimes; We resell a SaaS email service

IPMI/iDRAC (Out-of-Band Management)

Hardware management interfaces that provide remote, out-of-band control of servers for power, console, and diagnostics.

Reboot the host via IPMI; Mount an ISO through iDRAC; Restrict OOB access to the management VLAN

IPv4 and IPv6

Internet Protocol addressing versions: IPv4 uses 32-bit addresses, while IPv6 uses 128-bit addresses to vastly expand address space.

Offer dual-stack hosting; Plan for IPv4 scarcity costs; Add AAAA records for IPv6

Kubernetes

An open-source orchestrator for automating deployment, scaling, and management of containerized applications.

Run web pods behind an Ingress; Use Helm charts for deployments; Autoscale pods with HPA

Let's Encrypt and ACME

A free certificate authority and the protocol (ACME) enabling automated issuance and renewal of TLS certificates.

Auto-renew TLS certs every 60 days; Use DNS-01 for wildcard certificates; Certbot implements the ACME protocol

Load Balancer

A service that distributes traffic across multiple servers or containers to improve availability and performance.

Use cookie-based session stickiness; Health checks remove unhealthy backends; L7 rules enable blue-green routing

LTV (Lifetime Value) and LTV:CAC

Lifetime Value estimates total gross profit from a customer; the LTV:CAC ratio measures acquisition efficiency.

Target LTV:CAC of 3 or higher; Expanding ARPU increases LTV; Reducing churn boosts LTV

Managed Hosting

A service model where the provider handles operational tasks such as patching, backups, monitoring, and SLAs.

Choose managed WordPress; Add a managed database service; Our managed plans include patching and monitoring

Migration

Moving sites or workloads between servers or platforms, often involving data sync, DNS changes, and cutover planning.

rsync files and dump/import the DB; Lower DNS TTL before cutover; Run a final delta sync to minimize downtime

MRR (Monthly Recurring Revenue)

Predictable subscription revenue normalized per month, a core financial metric for hosting providers.

MRR grew 8% this quarter; Track new, expansion, and churned MRR; MRR forecasting drives hiring plans

Multi-tenancy

Hosting multiple customers on shared infrastructure while enforcing isolation, fair resource usage, and separate billing.

Isolate tenants with namespaces; Use cgroups to limit noisy neighbors; Per-tenant quotas and billing apply

Nameserver

An authoritative server that answers DNS queries for a specific domain (zone), serving its records to resolvers.

Register nameservers at the registrar; Use ns1 and ns2 for redundancy; Point your domain to our nameservers

NGINX

A high-performance web server and reverse proxy commonly used for load balancing, caching, and serving static content.

Terminate TLS at NGINX; Cache static assets with NGINX; Use NGINX as a Kubernetes Ingress controller

NVMe SSD

Solid-state storage using the NVMe interface for very high IOPS and low latency compared with SATA SSDs or HDDs.

Switch to NVMe-based instances for higher IOPS; Benchmark with fio; NVMe reduced request latency

Object and Block Storage

Object storage holds unstructured data as objects in buckets; block storage provides raw volumes for filesystems and databases.

Store uploads in S3-compatible object storage; Attach block volumes to VMs for databases; Lifecycle policies tier old objects

PCI DSS

A security standard outlining controls for any entity that stores, processes, or transmits payment card data.

Keep cardholder data off shared hosting; Run quarterly ASV scans; Tokenize payments to reduce PCI scope

Peering and Transit

Methods of Internet connectivity: peering exchanges traffic between networks (often settlement-free), while transit provides paid upstream connectivity to the full Internet.

Peer at an IX to cut transit costs; Use multiple transits for redundancy; Maintain IRR and PeeringDB entries

Plesk

A commercial hosting control panel similar to cPanel, widely used for Linux and Windows hosting management.

Manage Windows hosting via Plesk; One-click SSL in Plesk; Resellers get sub-accounts in Plesk

Registrar vs Registry

Registrars sell domains to customers; registries operate the authoritative databases for each TLD.

Register the domain at a reputable registrar; The registry for .org is PIR; Use EPP to transfer between registrars

RPO and RTO

Recovery Point Objective is the acceptable data loss window; Recovery Time Objective is the acceptable time to restore service after an outage.

RPO 15 minutes via frequent backups; RTO 1 hour using warm standby; Define RTO/RPO per service tier

Shared Hosting

Multiple customer sites share the same server resources, offering low cost but limited isolation and performance.

Entry-level plans are shared hosting; Resource limits prevent abuse; Upgrade to VPS when you outgrow shared

SLA (Service Level Agreement)

Contractual commitments on service uptime, performance, and support response, often with credits for violations.

99.9% uptime SLA with credits; Response time SLAs for support tickets; SLAs exclude scheduled maintenance

SSL/TLS

Cryptographic protocols that secure data in transit, most commonly used to encrypt HTTP as HTTPS.

Enforce HTTPS with HSTS; Use TLS 1.2+ and modern ciphers; Enable SNI for multiple certs per IP

TLDs (gTLD/ccTLD)

Top-Level Domains are the highest level in DNS, including generic TLDs (gTLDs) like .com and country-code TLDs (ccTLDs) like .uk.

Consider a ccTLD for local presence; New gTLDs like .app and .dev; Review registry policies before choosing a TLD

TTL (Time To Live)

The caching duration for DNS records that resolvers should honor before re-querying the authoritative nameserver.

Set a low TTL before IP cutover; Use 300-second TTLs for agility; Higher TTLs reduce resolver query load

Uptime and the nines

The percentage of time a service is available. Higher nines (e.g., 99.99%) demand more redundancy and operational rigor.

Five nines allows about 5 minutes downtime per year; Track uptime with external probes; Design for failure to meet uptime goals

VPC (Virtual Private Cloud)

A logically isolated virtual network in the cloud where you provision subnets, route tables, and security controls.

Create public and private subnets per tier; Use security groups and NACLs; Peer VPCs across regions

VPS (Virtual Private Server)

A virtual machine with dedicated resources on a shared physical host, offering more isolation and control than shared hosting.

Upgrade from shared to a VPS for more control; Resize VPS CPU and RAM as needed; Root access is available on the VPS

WAF (Web Application Firewall)

A security filter for HTTP(S) traffic that detects and blocks common web attacks and malicious bots.

Block SQL injection and XSS at the WAF; Enable bot management rules; Use a managed WAF in front of the origin

WHOIS

A protocol and set of databases providing public information about domain registrations, typically via registrars or registries.

Check WHOIS to confirm domain ownership; GDPR redaction affects WHOIS output; Use RDAP as the modern WHOIS replacement