Hey y’all!
I’m in the process of updating my nuxt app in order to use the edge rendering feature of Netlify.
I’ve successfully upgraded to Nuxt-Bridge-Edge.
However, when I test my legacy site against my new site deployed using Netlify, I get some very disappointing results.
Click here for the netlify report.
Click here for the digital ocean report.
Are there things I should know when using Netlify to speed up the initial response time? Any best practices I should be aware of?
Hi @Olivier_Lambert,
As far as I’m aware, Nuxt 2 was mostly static and now you’re moving to server-side rendering, correct? Did you try using same static export on Netlify? While, I’m not sure if that will give you performance better than DO or not, it would most likely cut down on response time. Edge Functions are meant to be fast, but that’s an additional piece of code that needs to be run before preparing your response and based on what’s happening in there, it could add huge latency.
Sorry for the confusion!
My Nuxt2 was running on SSR, so I couldn’t deploy on Netlify. The Nuxt2 App is hosted on Digital Ocean.
I was hoping that using Netlify edge deployment would lower my server response time as the server would be much closer to the client.
However, it doesn’t look too good. There must be something I’m doing wrong. Here’s the same build hosted on Netlify vs Digital Ocean:
Nuxt Bridge hosted Netlify
Nuxt Bridge hosted on Digital Ocean
By the way, the link that says “Deploy Nuxt 3 to Netlify doc” on this page is dead.
I was under the impression that Nuxt would, under the roof, be using edge functions where possible. Since a regular Digital Ocean appears to be faster, I assume that I’m wrong.
Is there any Nuxt related documentation I can sift through to better understand how to make it work?
Hey @Olivier_Lambert,
I don’t think the two reports are identical. The Netlify report you’ve shared is loading at least 0.5 MB extra data with 3 times more network requests as compared to Digital Ocean (120 on Netlify vs 40 on Digital Ocean). Again, I’m not justifying the speed difference, just that the reports seem to be incomparable.
I’d like to mention that putting raw numbers against one another is not a good way to compare, especially when there are many moving parts under-the-hood.