Stack Glossary
Web hosting — Where your page actually gets served from
Your site feels slow and you cannot yet say whether the theme, the plugins or the box underneath is responsible.
The short answer
Web hosting is the machine and the software stack that answers a browser request: a web server, a PHP runtime, a database, and a cache sitting in front of all three. Every number a speed tool reports about your site begins there.
The useful split is this — hosting sets your time to first byte, and your theme, images and third-party scripts set most of what happens after it. One request tells you which half you are fighting, and that single measurement saves weeks of guessing.
On our plans the stack is NVMe storage behind a LiteSpeed cache, current PHP 8.x with OPcache and MySQL, identical from the entry Sprint plan upwards.
By the HostingFast team · Reviewed 24 August 2026
100+
Terms defined here
2 min
Typical read time
Plain
English, no hand-waving
24/7
Engineers on shift
Strip the marketing away and a hosting account is a directory on a disk, a share of one machine's CPU and memory, a PHP runtime allowed to execute your code, and a database instance. A request arrives, the web server decides whether it can answer from cache, and only if it cannot does PHP run and MySQL get queried before any byte leaves the building.
That sequence is why two accounts on identical specifications behave nothing alike. A cached response leaves in single-digit milliseconds; an uncached one pays for every query the page makes. Knowing which of the two you just measured is the difference between tuning a site and redecorating it.
The request path, in order
DNS resolves the name, the browser opens TCP and negotiates TLS, and the request finally reaches the web server. LiteSpeed checks its cache first. On a hit, the rendered HTML is handed back from memory and nothing else on the machine is disturbed. On a miss, a PHP process boots your application, MySQL answers however many queries the page needs, and the result is assembled and returned.
You can watch that decision being made. Send a second request to the same URL and read the response headers: x-litespeed-cache: hit means the page never touched PHP, and a miss means it did. Almost every argument about hosting speed is really an argument about how often that header says hit.
Take the measurement before the opinion
Two numbers are enough to start. Time the server with curl -w against a URL that is already cached, and read the field Largest Contentful Paint in PageSpeed Insights, which comes from real Chrome users rather than a simulated run. The first is your host. The second is mostly your front end.
A cached page consistently answering above roughly 600ms points at the platform rather than your code, particularly if it flies at 7am and drags at 4pm. That pattern is contention — too many accounts competing for one machine — and no image compressor or caching plugin has ever fixed it.
What actually moves the number
In rough order of effect: caching at server level rather than in a plugin, a current PHP release with OPcache warm, fewer and better database queries, and images served at the size they render. Everything else is decoration. Stacking three caching plugins on top of a server cache is the classic cargo-cult fix, and it usually makes the site slower and much harder to debug.
Change one thing, then re-measure the same URL at the same hour. If the number did not move, put the change back. A tuning session with no before-and-after column is not tuning; it is a list of things you now have to support forever.
What runs underneath HostingFast
Every tier gets the same foundation: NVMe SSD storage behind a LiteSpeed cache, current PHP 8.x with OPcache, MySQL, HTTP/2, and a London datacentre with redundant power and several upstream carriers. For a British or Irish audience that is 5 to 15 ms of network round trip before your server has done anything at all.
The neighbouring entries worth reading next are Shared Hosting for what density does to that stack, TTFB for how to time it properly, and LiteSpeed for what the cache is actually doing when it says hit.
Definitions with the mechanism left in
Most hosting glossaries stop at the analogy. This one keeps going to the part you can act on: what the component does on the request path, which command shows it working, and what a good reading looks like.
Free migration means you can measure honestly. Baseline the old host, let our engineers move the site, then run the identical test again and see the delta rather than trust a claim.
- The request path, component by component
- The command that proves each claim
- Numbers worth aiming at, not slogans
- Written by the engineers on the tickets
Why HostingFast
Standard on every plan
NVMe on every tier
Storage is NVMe from the entry Sprint plan upward, so a database read never queues behind a spinning platter.
LiteSpeed ahead of PHP
Rendered pages come out of memory. A repeat request returning x-litespeed-cache: hit is the proof, and it costs nothing to check.
Current PHP 8.x with OPcache
Compiled bytecode is held between requests, and the release is set per site from the panel rather than per server.
A migration you can measure
Our engineers move the site free of charge, which means a clean before-and-after on the same URL rather than a promise.
A real shell, not a file manager
SSH with Git and Composer on shared plans, so WP-CLI runs and deploys happen the way your project already expects.
TLS that renews itself
Free SSL is issued once the domain resolves here and renews ahead of expiry, so no certificate cron of yours can quietly fail.
Quick Start
From order to online
- 1
Baseline the site you already have
Time a cached URL with curl -w and note the field LCP from PageSpeed Insights. Two numbers, written down, before anything changes.
- 2
Order and hand us the migration
Sprint is billed annually and goes live minutes after checkout. Your old host keeps serving visitors until DNS moves, so the copy is verified before anything is switched.
- 3
Re-run the identical test
Same URL, same hour of day, same tool. Server time should move first; front-end numbers only shift once you change the front end.
Built In
Loaded onto every plan
- NVMe SSD storage behind a LiteSpeed cache on every tier
- Current PHP 8.x with OPcache, selected per site
- SSH with Git and Composer on shared plans
- Free migration carried out by our own engineers
- Free SSL issued and renewed without a reminder
- Daily backups, restored by you from the panel
- 240+ applications in the Softaculous installer
- 99.9% uptime as the target, monitored around the clock
- First reply targeted at two hours, one when a site is down
- Renewal billed at the rate you first ordered at
Frequently Asked
What people ask us most often
How do I tell whether the host or the site is the slow part?
Time a cached page from the command line and compare it with the field data in PageSpeed Insights. A quick server response with poor Largest Contentful Paint is a front-end problem: images, fonts, third-party scripts. A slow server response on a page that should be cached is the platform, and no plugin will touch it.
What should time to first byte look like on a shared plan?
For a cached page served from London to a UK visitor, the network alone costs 5 to 15 ms and the cache hit adds very little on top. The figure worth watching is stability rather than a headline: if a cached URL answers quickly at 7am and slowly at 4pm, you are measuring contention on the machine, not your code.
Will a bigger plan fix a slow page?
Only if you have proved the page is CPU or memory bound. More cores do nothing for a page whose time is spent in twelve uncached database queries or a 4 MB hero image. Measure first; the upgrade that helps is the one you can already justify with a number.
Can I use WP-CLI, Git and Composer without a VPS?
Yes. Shared plans include full SSH access with Git and Composer available, so wp cli commands, composer install and git pull all work from a normal terminal session. That is usually enough to keep a small team deploying properly without taking on server administration.
Who operates HostingFast?
HostingFast is a trading name of Vitalcare at Home Ltd, registered in England and Wales, and the terms you agree to are governed by English law. Checking the company behind a host takes two minutes on the public register, and it is worth doing before you point a domain at anyone.
Keep reading
VPS (Virtual Private Server)
Guaranteed CPU and memory with root access, and what that actually changes about your response times.
Shared Hosting
One machine, many accounts, and the density question that decides whether shared hosting is quick or not.
Best LiteSpeed Hosting
The LiteSpeed buyer's guide, with the header check that proves a host is really serving from cache.
WordPress Hosting
Managed WordPress on NVMe with LiteSpeed and Redis, staging copies and daily backups included.
Magento Hosting
Magento wants memory and PHP headroom; these are the tiers that supply both without an argument.
Changing hosts? Run through our checklist first.
A straightforward sequence for a switch your visitors never feel: which files move first, how to shift email across without losing a single message, the right moment to repoint DNS, and the two mistakes behind almost all the downtime we get asked to rescue.
Move it, then measure it.
NVMe and LiteSpeed on every tier, with migration done for you — baseline your server time before the move and again after.
View WordPress Hosting plans