Is 100 or less requests per second (for non-cached pages) what one can expect with Rails?

Question

Preface: Please don't start a discussion on premature optimization, or anything related. I'm just trying to understand what kind of performance I can get from a single server with rails.

I have been benchmarking ruby on rails 3 and it seems the highest rate of requests per second I can get is around 100 requests per second.

I used phusion passenger with nginx, and Ruby 1.8.7.

This is on a ec2 m1.large instance:

7.5 GB memory
4 EC2 Compute Units (2 virtual cores with 2 EC2 Compute Units each)
850 GB instance storage
64-bit platform
I/O Performance: High
API name: m1.large

The page was a very simple action that wrote a single row into mysql.

user = User.new
user.name = "test"
user.save

I am assuming no caching (memcache, etc), I just want to get a feel for the raw numbers.

I used apache bench on the same ec2 instance, and I used varying levels of # of requests (from 1000 to 10000 and varying numbers of concurrent requests 1/5/10/25/50/100).

Answer 1

The EC2 m1.large instance is really not that fast, so these numbers aren't surprising. If you want performance, you can either spring for a larger instance, as there are now some with 80 ECUs, or try a different provider.

I've found that Linode generally offers higher performance for the same price, but is not as flexible and doesn't scale to very large numbers of servers as well. It's a much better fit if you're in the "under 20 servers" phase of rolling out.

Also don't think that MySQL is a no-cost operation.