Big website architecture question

For a website that should allow searching based on a few filters on about 100 millions records, is it good enough to pick Solr for searching?

The website would be written in C#, (asp.net mvc), the backend database with 开发者_StackOverflowuser data and shopping cart should be SQL Server 2008R2. Does it scale ok for 10 million users? Other stuff that should be used: Memcached, HAProxy probably...

I have not directly worked with solr, but it is highly scalable and I have worked at a company that used it for asp.net webform sites wtih extremely high traffic (e.g. a national newspaper, online magazines). Yes, it is based in Java so there would definitely be a mismatch in technology, but it scales very well with replication and 'sharding' technology. see wikipedia. It's also fairly stable - it was created in 2004. You can see a number of very large traffic sites that use it here

You haven't received any answers since the question is a bit vague.

It all depends on what the records contains and what your "filters" are.

The best way is really to create a program than generates millions of record and then try to do some searching.

Another solution is elasticsearch, which is also based on Lucene