Thread

Hi EE forums!

I was hoping to get some extra brainpower and advice on the best way to execute a project we’re working on. Here’s the rundown:

1.8M professionals
10 different services
44K zip codes

We’re working on developing a database of professionals, so that the user can enter their ZIP code, a desired service and see all matching professionals. We’re concerned about performance and interested in advice on the best way to execute this.

Each professional may serve multiple zip code, and offer multiple services. We thought about trying to setup categories for each zip code and service, and associating each professional entry with the appropriate categories—but we’re worried that ~44K categories will choke EE.

Some related modules we thought might be helpful:

http://devot-ee.com/add-ons/geotagger
http://devot-ee.com/add-ons/geofinder
http://devot-ee.com/add-ons/playa
http://devot-ee.com/add-ons/us-states-cities

. . .

Hopefully it’s clear what we’re trying to accomplish. We’re flexible... if there’s a smarter way to accomplish this, please don’t be shy to go into detail. Any help or advice on the best way to go about this and the server/hardware needed would be greatly appreciated!

Thank you

Hmm..

First thought is that sounds like requirements for a custom app. That’s a pretty high expectation of a $300 CMS.

I’m no MySQL expert so can’t speak to things at that level, but I can’t imagine any part of EE’s Control Panel scaling that big - if the database scheme can support it you’ll be on the hook for a custom backend UI to work with that number of entries and categories.

Will all of that content be in the system right away? Or can you use EE as a quick prototyping tool to prove need/profitability and then migrate to a custom app when the time is right?

Hi Boyink!,

First off, thanks for taking the time to reply. I appreciate it.

Will all of that content be in the system right away? Or can you use EE as a quick prototyping tool to prove need/profitability and then migrate to a custom app when the time is right?

Using EE as a proof of concept for let’s say for 2-3 states, and then migrating to a custom app when the time comes to support the entire US is certainly a potential option. If we took that route however, it would still be about 100K entries and potentially 10K zip codes if we supported all professionals in the area; we were thinking of using the “tags” module and entering the zip code as a tag rather than having thousands of categories, in hopes of this performing better.

. . .

Maybe it’s better to approach this from the other end—what is the maximum entries, categories and tags EE can handle on a typical VPS or dedicated server? Finding a solution will be easier if we know what we’re working with.

Thanks!

How many angels can dance on the head of a pin? 😉

An entry != an entry - so much depends on the number of and type of custom fields in the channel and total number of channels & fields (knowing that each custom field adds 2 fields to the channel_data table).

Not to mention the coding & caching practices used on the front end.

Maybe taking a look at http://www.hopstudios.com/blog/the_largest_expressionengine_sites/ and contacting some of those sites might help.

Maybe taking a look at http://www.hopstudios.com/blog/the_largest_expressionengine_sites/ and contacting some of those sites might help.

Thanks Boyink!

I’ll see what I can uncover. I hope you’ll still be around in a day or two, I may have more need for your expertise. Again, I appreciate you taking the time to help.

p.s.

How many angels can dance on the head of a pin?

42 😝

Hello nModo,

I am glad that Mike was able to offer you some help.

If you need anything else, just let us know!

Cheers,