Time to first byte/wait time high

July 20, 2015 3.1k views
Server Optimization MySQL

Hi,

I migrated a site that was previously hosted elsewhere to a 4 CPU droplet.
The wait time more than doubled here on a droplet.
The site is slower to browse, it's even obvious for a human being, even though I use a CDN, minify HTML/CSS, use other cache solutions like I did on the previous server. Google Speed Insights showed the time to connect as an issue which wasn't the case on the previous server (which was running on an older CPU with HDD).
GT Metrix waterfall show a wait time of over 800ms on average, while it was less than 300ms before.

I come from a dedicated server, so maybe that's one of the cons with a VPS but I wasn't expecting such a difference.

CPU and RAM are consistently low so that isn't the problem either.
I use Server Pilot as control panel.

Any help would be appreciated.

Thanks

3 Answers

It is odd that with caching you'd still see such a high time to first response. The most common reason for this in Apache is the KeepAlive and KeepAliveTimeout settings. If the timeout is too high and you have KeepAlive enabled, apache may not be able to respond to additional requests until the already open sockets have timed out.

We to have found the same.
Subscribing in case anyone ever answers in the future.

This is a famous problem with DO.

To test whether it is due to DNS, PHP, Wordpress, hosting or some other issue, I duplicated my website in Google Cloud using the Google Compute Engine instance.

Problem solved! I'm now able to get faster server response times and pass the Pagespeed test with flying colors even with the smallest, cheapest VPS box.

So it's not due to PHP or Wordpress. Digitalocean server does have server response time issues. With Google I can get 100-300ms response time, whereas here it's more like 1.0-1.2s.

But comparatively, DO is cheaper and has nicer UI :)

Hope this helps.

Have another answer? Share your knowledge.