Best ecommerce solution for a big site?

best app for big ecommerce site

[rolygate]

I realise this is of course a personal question but would appreciate some 'votes'. I have to project manage a large implementation soon and need to canvass opinions. Unfortunately my experience is limited to sites with less than 15K products. Here's the outline spec; I would be equally happy with 'no' votes, as well.

The benchmark is Miva, which is capable enough (100K products +), the right price, safe, stable, secure and has plugins for most purposes; but is definitely not popular with our team, for working with (proprietary code, non-optimal templating, support issues, etc).

- Must be proven to be capable of 50K products +.
- Must have big livesites for proof.
- Must run on LAMP servers.
- Can't be a megabuck solution.
- Database: either mySQL or postgreSQL.

...and the wishlist:
- More developer support than Miva.
- Real template solutions.
- Native short URLs capability.
- Code that validates.

Many thanks in advance for your opinions. Please do make suggestions even if you are not sure that your suggested app is 100% compliant with my spec. OSS is perfectly acceptable if the app and support have the necessary reputation.

[intaglio]

I probably wouldn't push for a LAMP solution. LAMP only makes sense if you're planning to use a shared web host.

I'd lean more for a well designed runtime framework that clusters well (jsp with a fast database backend).

Either way be prepared to spend some major big bucks to get it up and running.

[Peat]

LAMP is stable and has broad community support, but "big" probably means several dedicated servers, so your options open up quite a bit -- Java, .NET, and Rails are all proven under heavy loads, and may have more of the tools you'll need for a complex environment.

I've only worked casually with Miva, and I've never heard anything good about it on a large scale. You'll be fighting an uphill battle, especially since big sites require extensive customization and integration with external services, which Miva is notoriously poor at.

All of your stated requirements can be achieved with any almost any development platform, and so I have the feeling that more research needs to be done:

* What external services and systems will you need to be interfacing with (inventory, logistics, etc.)?

* How many people, in what roles, will be interacting with the system?

* What are the business goals for the next few years?

* Do you have any immediate problems that need addressing?

These are the sorts of questions that will determine what technologies are right for your project.

My recommendation is to spend a little time and money up front putting in the proper research into the business requirements for their next generation site. These technical questions are fun to ponder, but without the proper guidance and goals, making early technology decisions will inevitably run over budget and frustrate the crap out of everyone involved.

Feel free to shoot me an e-mail if you want to discuss it in more depth; I'm always keen to learn more about these sorts of projects.