Whizlabs scalability question

[Mika Tapanainen]

Hi guys,

I found in Whizlabs a question and I'm not completely sure about the explanation they gave.

In the question was a 2-tier application where the presentation logic and business logic were in the same tier in the web server and the database in the own server.

There was that there is a potential db access bottleneck, because every client needs own connection for the database. Is here the reason that in the 3-tier architecture the connection pooling is possible and in the 2-tier architecture not?

Thanks