Hi!
My name is Tomasz Kustra, and I am from Poland.
I am interested in this project.
A moment ago was involved in another project about amazon scraping.
The main goal was to scrape reviewers data.
It was big job, resulting in over 60mln products and reviewers info.
those projects:
https://www.freelancer.com/projects/postgresql/Fix-Data-Scrapping-Perl-Script/
https://www.freelancer.com/projects/php/Project-for-tomkusvw-9887178/
Unfortunately there is no indication that it was amazon :(
two smaller projects have:
https://www.freelancer.com/projects/php/Amazon-grabber-list/
https://www.freelancer.com/projects/php/need-someone-can-build-kind/
Your project will become a huge one, because from one product you will get few more, from different categories, so you will finally gathering millions of products.
It will take much time, too.
For this you will need using proxies and multi-threading (without it you will be very limited by time of getting page and decaptching - to no more than average 2 product page per second, so to get 20 mln products it will take more than 3 month.)
I was using 50 proxies and over 40 threads, but number depends on server power.
Of course such projects are very customized, but it won't be a problem for me to customize the code and get you all data.
I was using PERL on linux and PostgreSQL for it.
I know that my bid is higher than average, but I know how big this task will be.
Contact me and we can discuss more.
Regards
Tomasz