I am using Postgre with NodeJS. If I fire of 500 threads to test the scalability of the application it fails. I have heard Postgre has limitations when it comes to connection capacity. I don't have this problem with SQL Server or MongoDB. Each thread makes a call to Node, which then calls a simple Sql Statement to pull back the data. However, It works fine for 5 threads. Again, I need to know how to scale Postgre so it is a viable option.
Have you done anything else besides fire off more threads than the default connections limit [1] and watch something fail somehow to conclude that it is not viable. People are successfully using PostgreSQL to handle considerably larger loads. Can it handle those "out-of-the-box"? No. The product takes time to learn and many people have posted many things, including the documentation, to aid in that education process. Searching for those and then asking specific questions, with supporting details, usually facilitates higher quality help.
Being open source you tend to get out much more than you put in. However, it seems to me that you need to input a bit more than you have here to get reasonable help quickly. In particular showing people exactly what failed by posting log messages.
Also, help comes more readily if you post your requests to it to the proper mailing list. Our bug reporting list is not intended for usage questions. -general is typically the better choice, though -admin or -novice both would have merit here. Admittedly those all require subscribing to the mailing list in order to post to them, but that's how we roll here.
David J.
P.S. The community really prefers if you use either "Postgres" or "PostgreSQL" when referring to our product. "Postgre" looks and sounds like slang.
