It depends on your definition of 'very large scale'. Certainly any large financial/insurance company is going to be churning through great gobs of data that really does have to be universally consistent. This is basically the niche that keeps IBM's mainframe division in business.

I'd suspect that if those don't qualify as 'very large scale' to your definition (which would be entirely justifiable) that such systems don't exist because they're not currently possible. For example I don't think it's possible to implement Facebook with SERIALIZABLE no matter how much money you throw at the problem.

(I'd love to be proven wrong of course)

I'd guess that in Amazon and other online retailers, the part that completes the order of the item from the warehouse has to be transactional so two customers won't buy the last item.

Normally, whenever physical objects or money is involved, transactions should be considered.