Calculating the throughput of a server

[baijuw]

I have read many articles on hdparam to calculate the disk read and write speeds and some on interface and CPU limits. But is there a structured way of calculating the maximum throughput of a server including all the subsystems. Like storage, CPU, network, memory and so on? So that I can create a script that i can run on a newly installed Linux machine and calculate the maximum throughput .

[Irithori]

adhoc and/or OneTime:
use (h)top, vmstat, iostat, dstat, fio

Monitoring:
munin, xymon, nagios, etc

[coopstah13]

thats kind of a pointless exercise

how much throughput you get is going to depend on the types of things running on your server and the load generated against those things

You should be testing load against your applications

[baijuw]

Thats right. running standard commands will only tell you how your system is performing at that given time.
What I am really looking for is a real test like the one which hdparam does to tell you how fast you can read or write data onto your hard drive. But this is limited to just HDD. What we should really be talking about is the cumulative throughput of all the subsystems . Now I do understand that throughput is usually expected for an application. So let us assume that it is an ftp site or a site like linuxforums which gets multiple inputs. how can one safely say that linuxforums can accept "x" number posts per second.
what kind of action should I perform on my server and study its behavior to arrive at a safe figure?

[coopstah13]

use a tool like grinder or jmeter to drive a web application running on your server

if you were running a forum and wanted to see the number of posts per second, you would write a test to keep posting over and over, increase the threads until you start to see a decline in throughput