When does a Wordpress site become too big and slow?

bernard

BuSo Pro
Joined
Dec 31, 2016
Messages
2,527
Likes
2,217
Degree
6
Does anyone know what is the upper limit for running a smooth WP site in terms of database size or number of products/posts?
 
If you run the DB on a separate server like AWS db instance, and do caching, it would be pretty damn high. Few million posts maybe?
 
If you run the DB on a separate server like AWS db instance, and do caching, it would be pretty damn high. Few million posts maybe?

Cool.

I was thinking about if you wanted to do a proper price comparison website, where you'd want to do parent-child relationships with products (child for each vendor).

Like importing multiple feeds and trying to match their products by some RegEx of product names or even image file names and such.

Would Wordpress even be able to do this properly or would it be too complex?
 
If you run the DB on a separate server like AWS db instance, and do caching, it would be pretty damn high. Few million posts maybe?
What are the benefits of running the DB on a separate server instead of the classic nginx setup? I've always wanted to know
 
Would Wordpress even be able to do this properly or would it be too complex?

Not sure what your programming background is but you could get a MVP up with WordPress but if it was me personally I would look to use something else other than WordPress for this use case.

Perhaps something like Laravel if you are already familiar with PHP or NodeJS, they leave you open to add more complexity as it arrives which it will from my experience building price checking websites and quicker. You can always bolt on a headless wordpress CMS for the blog section of your website to keep a familiar CMS for freelance writers etc.
 
Not sure what your programming background is but you could get a MVP up with WordPress but if it was me personally I would look to use something else other than WordPress for this use case.

Perhaps something like Laravel if you are already familiar with PHP or NodeJS, they leave you open to add more complexity as it arrives which it will from my experience building price checking websites and quicker. You can always bolt on a headless wordpress CMS for the blog section of your website to keep a familiar CMS for freelance writers etc.

Yes, it's because I already have a Wordpress site and don't really want to port it yet.

The new site I'm doing, in a different language but similar, is what I will likely do in Django (I code Python).

For now, I think I will create a database (MongoDB?) with Python and then do harmonizing of the feed data into it, then export it as XML with Custom Post for each product and then use custom fields for vendor prices and even repeater ACF fields for historic price data. Import with WP All Import.

It is just so ugly, but pragmatically I think that is what I need to do right now.
 
Checkout the rehub theme and is plugins, I think it supports vendor/product relationship out of the box.
 
Some of the biggest sites on the net are run on Wordpress and you'd never know if you didn't dig into the source code and look for footprints. Imagine how bloated they are with 10 revisions of 100,000 draft articles each, plus a million posts published, with 4,000 user, guest contributor, editor, and admin accounts, and got knows what else they're tracking in there. And they all run great.

As long as you're using server-side caching and not expiring and re-caching old posts that will never change, there should be minimal problem. Even without caching, Wordpress does a great job of indexing where everything is so the queries are fairly optimized.
 
@Ryuzaki do you have any opinion on building a product comparison database from feeds and scraped info?

Should I do a parent/child relationship, where the parent is the product placeholder and the child is the individiual products from vendors, or should I use one product post and then add each vendor product info as custom fields?
 
@bernard, I think either architecture style could work but I think you'd be more organized with the 2nd style with one page with all of the products as custom fields.

I'm trying to think about whether or not it'd be easier to loop through the data to show the products you need using the parent page / child page setup. I think it should be about the same amount of difficulty. I'm assuming you're talking about Advanced Custom Fields here, which will make it easier both in building out the architecture during the imports and looping through it all.

With the disconnect between the way you're getting this done and the final product, I'm doubting you'll ever really go in manually and edit single fields or anything, so talking about being able to filter through child pages versus doing a "find" search on the one page isn't that big of a deal.

I'd say whatever makes sense in your own mind and keeps you sane is best. I don't see any "efficiency" differences in querying or anything like that. It might be easier to keep it all using ACF functionality for the most part than mixing in more vanilla Wordpress stuff too. But once it's built, it's built, so I'd go for whatever is comfortable and makes sense in my own mind.
 
Back