What’s totally missing from your comparison and performance assessment is the traffic or LOAD on the respective sites/servers.
For the same site, a server that is just enough to power the site when you have just 1,000 visitors per month will not be able to power the same site when you have a million visitors per month.
And a 1-page site that gets a million visitors per month (eg if you’re driving paid ads to it) will require a much bigger server to run than a website with 100,000 posts that was launched today and gets virtually no traffic.
That’s why site owners often scramble around to beef-up their server when they know they’re going to appear on a major news site: because the huge inrush of visitors will put excessive strain on their “normal” server, crashing the site. (The Slashdot effect)
For a 5-year-old site with 82,000 posts, I can bet it’s getting a ton of traffic/visitors, and this is playing a big role in the slow dashboard speed you’re experiencing. Heck, even bots alone can put a huge strain on the server (if bot crawling is not properly controlled).
I have two site both use the same theme and identical plugins also both hosted on Amazon AWS Lightsail
These plugins (and the theme) aren’t just sitting there idle: they are doing “something” (that’s why you installed them to begin with). And to do this “something” on 82,000 posts for (say) 100 visitors simultaneously will require much more infrastructure than doing the same “something” on just 18 posts for (say) 10 visitors.
1st site 5 Years old – https://bit.ly/3k9mKMz – 82000 posts
– Instance set in Ohio USA (I am in US)
– 32 GB RAM, 8 vCPUs, 640 GB SSD Bitnami WordPress Instance
– 2 GB RAM, 1 vCPU, 80 GB SSD MySQL database (8.0.21)
I’m really struggling to understand why you have such a configuration. For a database-driven website like WordPress, most of the load (hence the bottleneck) will be on the database.
So I’m really perplexed why you have a MySQL database server with just 2GB RAM and 1 vCPU, while the webserver rather has 32GB RAM and 8vCPUs. Did you choose these RAM and CPU configs based on expert analysis of the performance logs on the two servers? I doubt this was the case.
In summary, yes, the number of posts play a role, but even more important is the load or traffic coming to the site: the 5yo site with 82,000 posts is likely getting traffic orders of magnitude higher than the 3wo site with just 18 posts. And comparing the performance of two websites as you’re doing now isn’t really useful.
And looking at the GMetrix report you provided, even the “winning” 3wo site with just 18 sites isn’t performing spectacularly: LCP of 5.4 seconds and an “F” GMetrix grade.
This leads me to conclude that your setup for both sites isn’t optimal, and you can probably get better performance without spending more, if you could have an expert look at your setup and tune things up for you.
Just my 2 cents.
Good luck!
-
This reply was modified 3 years, 8 months ago by George Appiah. Reason: Fixed typo