Performance Glossary
The 301 Redirect, Measured in Round Trips
Your migration left redirects behind, first-byte time crept up, and nobody can say how many hops a request now takes before the HTML even starts.
The short answer
A 301 permanently forwards one URL to another and carries ranking signals with it — and for anyone watching a waterfall, the fact that matters is that each hop is a complete round trip finished before the browser has asked for the real page.
So the useful question is never whether to redirect. It is how many hops fire, and which layer serves them. One curl invocation answers the first; moving the rule out of PHP and into the web server answers the second.
By the HostingFast team · Reviewed 12 August 2026
100+
Terms, measured not asserted
2 min
To read one entry
Plain
Commands you can run
24/7
Engineers on shift
Signals travel through a 301, which is why migrations lean on them. Latency travels through as well. A hop that stays on the same host reuses the open connection and costs roughly one server round trip; a hop that changes host repeats the DNS lookup, the TCP handshake and the TLS negotiation from scratch.
The 302 is the temporary form and deliberately withholds the permanence signal. It also behaves differently in intermediary caches, which on its own is reason enough to stop reaching for it on moves nobody intends to reverse.
Count the hops before you argue about them
One line settles it: curl -sIL -o /dev/null -w '%{num_redirects} %{time_total}' against the old URL prints the number of hops and the wall-clock cost of the whole chain. One hop is normal. Two is a smell. Three usually means two people wrote two rules and never compared notes.
The browser tells the same story if you tick Preserve log in the Network panel first. Each hop appears as its own row, with its own 301 status and its own timing bar. Without that tick the rows disappear the instant the destination paints, which is exactly why so many chains survive for years.
Same host, or a whole new connection
From our London platform a UK visitor sits roughly 5–15 ms away and a visitor on the US west coast 130–150 ms away. A same-host hop costs one of those round trips. A cross-host hop adds a fresh DNS lookup, TCP handshake and TLS negotiation on top, which is the difference between a rounding error and something a person on a phone can feel.
The bare-domain-to-HTTPS hop is the one worth attacking first, because it sits on the critical path for anyone who types the address. Served from .htaccess it is handled by LiteSpeed without waking the interpreter; served by a plugin it costs a full application boot to emit one header.
Flatten the chain, keep every rule
Chains assemble themselves. A migration writes /old to /new, an HTTPS rule bolts on in front, a trailing-slash rule joins the back, and nobody ever decided that A should reach D by way of B and C. Repoint each source straight at the final destination and the chain collapses to a single hop with no loss of coverage.
Never shorten a chain by deleting its oldest rule. Links on other people's sites are not updated, and the traffic they send is the cheapest you will ever get. Keep the rule; shorten the path it takes.
Proving the flattening actually landed
Take a before figure with that same curl one-liner across your ten busiest legacy URLs, apply the flattening, purge the cache, then run the identical command again. The hop count should fall and time_total should drop by about whatever the removed round trips cost on that connection.
Then wait for field data. This is a first-byte improvement, so it surfaces in real-user TTFB and in the LCP of pages people reach through old links — not in a lab run against the destination URL, which never saw the chain in the first place.
Definitions written by people who profile things
Every entry here is written the way we would answer it in a ticket: the mechanism first, then the command that measures it, then the number that tells you whether what you measured is a problem.
Moving in from somewhere slower costs nothing. Our engineers bring the whole site across, normally inside 24 hours, and the old one keeps serving visitors until the new one is ready to take the traffic.
- Mechanism first, tip second
- A command you can paste and run
- Thresholds quoted, never invented
- Redirect chains counted, not guessed
Why HostingFast
Standard on every plan
The mechanism, not the mantra
Each entry says what the machine is doing, so the next problem is reasoned about rather than pattern-matched.
A command attached to the term
Where something can be measured, the entry names the tool and the exact invocation to run.
Numbers with a source
Thresholds come from published Core Web Vitals guidance or from our own platform, never from a marketing deck.
301 chains, hop by hop
What one redirect costs, how to count a chain, and which layer should be answering it.
Failure modes named
How each thing breaks under load, and the reading in your tooling that says it has.
Wired to what it touches
Terms link to the ones sharing a bottleneck, so one lookup ends up explaining three readings.
Quick Start
From order to online
- 1
Export the URLs that still redirect
Pull the 3xx rows out of your access log for the last week and sort by hits. That list, not the rule file, is what your visitors are actually walking through.
- 2
Run curl against the top twenty
Record num_redirects and time_total for each. Anything above one hop goes on the fix list with its cost written next to it.
- 3
Rewrite sources to point at the destination
Edit each rule so it targets the final URL, purge the cache, then re-run the same curl loop and compare the two columns.
Built In
Loaded onto every plan
- LiteSpeed compiled into the server, not a caching plugin bolted on afterwards
- NVMe on every tier — the entry plan runs the same drives as the top one
- PHP version selectable per site, switched from the control panel in seconds
- SSH, Git and Composer on the developer-focused plans
- Staging you can clone, break and throw away before anything reaches live
- Daily backups with a self-service restore you run yourself from the panel
- Free SSL on every plan, reissued automatically well before it can expire
- Free migration by our own engineers, normally inside 24 hours
- 99.9% uptime as the target, monitored around the clock
- Renewal billed at your signup rate — no year-two step change
Frequently Asked
What people ask us most often
How long before a 301 is fully reflected in search results?
Crawlers see it on the next fetch; the consolidation of signals onto the destination takes several weeks. That timing is not a reason to treat the rule as temporary — external links never get updated, so the redirect earns its keep for years. Measure the traffic on the destination URL, not the crawl date.
Should the redirect live in .htaccess, in PHP, or in the CMS?
As far up the stack as it will go. LiteSpeed answers an .htaccess rule without starting PHP at all; a redirect plugin boots WordPress, loads the plugin stack and hits the database before it can send a header. On a busy legacy URL that difference is the entire response time.
Where do the servers sit, and what does that cost me in latency?
London, with redundant power, cooling and several upstream carriers. Distance sets the floor on first-byte time — roughly 5–15 ms for a UK visitor and 70–90 ms from the US east coast. Past that floor the engineering decides everything: NVMe storage, LiteSpeed caching and restrained account density per machine.
What does a migration actually involve at your end?
You hand over the login for your current host and our engineers move files, databases, mailboxes and configuration across, normally inside 24 hours. You test the copy on our hardware before DNS changes, so the old site keeps answering visitors right up until the new one is ready to take over.
Keep reading
Domain Name
The readable address DNS resolves, and the lookup that precedes every first byte.
.htaccess
Per-directory rules read on every request, and the ones worth moving upstream.
How to Set Up 301 Redirects
Writing the rules, flattening the chains and checking the result with curl.
Ecommerce Hosting
Storefront hosting built around the pages a cache cannot touch: basket and checkout.
OpenCart Hosting
One-click OpenCart on storefront-tuned PHP, with free SSL and staging included.
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 the redirect rules onto faster ground.
NVMe storage, LiteSpeed and free migration on every plan, with a renewal price that does not move on you.
View Ecommerce Hosting plans