What we do · Data and web
Web application performance audits
When a site is slow, the easy answer is "add more servers". Ours is to measure every tier, find the one thing actually holding requests up, and fix it, with numbers before and after. Often the servers turn out to be idle and the cause sits somewhere nobody was looking.
What it is
A structured investigation of a slow or unstable web application, from the outside in. You receive a branded report that names the root cause tier (infrastructure, application or third party), shows the evidence for each tier, and lists fixes in priority order. Where the fix is ours to apply, we apply it and measure again.
We cover WordPress, Magento, PrestaShop, Laravel and custom PHP, on plain Linux, Plesk and cPanel with CloudLinux, behind nginx, LiteSpeed or Apache.
How we build it
Infrastructure first, to rule it in or out. We prove whether the servers are busy or idle while pages are slow. If CPUs are mostly idle and the database is answering from memory, adding hardware will not help, and we say so.
Outside-in timing. We time requests through the CDN, at the edge and at the origin, cached and uncached, so we know where the seconds are spent before reading a single log.
PHP, by blocking frame. PHP-FPM slow logs are grouped by the function each slow request was actually stuck in. That turns "the site is slow" into "forty percent of slow time is one external call".
Database, by evidence. Slow query logs with EXPLAIN plans, lock analysis and buffer pool behaviour. We size the InnoDB buffer pool to the working set, correct per connection buffers that are set too large, and find the unbounded queries and table scans that actually burn CPU.
Every external dependency. Payment gateways, ERP and stock APIs, licence servers, fonts and analytics: we probe each one, because a synchronous call to a slow third party stalls the page as surely as a slow query.
Fixes that hold. Pool sizes set from measured worker exhaustion, request termination timeouts and memory ceilings so one bad request shape cannot take the pool down, and nginx and Varnish buffer limits that fit real cookies and headers.
What it protects you from
- Paying for more hardware when the hardware is not the problem.
- One misconfigured database setting turning a healthy server into a CPU or storage I/O fire.
- Years of scheduled restarts standing in for a root cause.
- Slow checkouts caused by a third party you did not know was in the request path.
- Arguments between hosting, developers and vendors with no data on the table.
In practice
A sports retailer's PrestaShop store. Thousands of requests a day were taking over four seconds while the web servers sat mostly idle. We attributed about 40 percent of slow time to a synchronous ERP stock call with a very long timeout, 24 percent to queries scanning millions of rows, and 12 percent to faceted navigation. The customer received a branded report with evidence per tier and a prioritised fix list for their developers.
Why EuroVPS
- We measure before we recommend, and we measure again after the fix.
- We read the whole stack: network, storage, database, PHP, web server, CDN and third parties.
- Reports name a cause and a tier, with evidence your developers can act on.
- We apply the infrastructure side of the fix ourselves, safely and reversibly.
- Honest outcomes: if your servers are fine, we tell you, even when selling hardware would be easier.
FAQ
Do you need access to our code? No. We work from server, database and access logs and from outside timing. If the cause is in the code, we point to the exact plugin, endpoint or query.
Can you audit a site hosted elsewhere? Yes, with the access needed to read logs and metrics. Many audits end in a migration; some do not need one.
How long does an audit take? Many causes are found in hours. A full report with every tier measured usually takes a few days, depending on how intermittent the problem is.
Will you change anything without asking? No. Every change is agreed, applied with a rollback path, and measured.
What do we receive? A branded report with the measurements, the root cause tier, and a prioritised list of fixes with owners.
Slow site, idle servers? Talk to an engineer.
Related services
Your next infrastructure partner
Let’s look at
your setup.
Tell us what you run today and what needs to improve. An engineer will review your brief and help you plan the next step.
+357 222 32280
who will look after your infrastructure.
Let’s start with your setup.
A little context helps us start the right conversation.