How many simultaneous requests can be handled by a medium class server on the average?

Question

I have bought a PRIMERGY TX100 S1 Server with a trial version of Windows Server 2008 R2 Web Edition.

My internet connection with a static IP is very very fast (about 50 mega bit per second) for both downloading and uploading.

My site serves text based contents only, no streaming.

How many simultaneous requests can be handled by a medium class server on the average? Can it handle at least 1000 simultaneous requests?

Edit 1

For those have experiences in this field, they might have a rough prediction.

Any more details like server specs, objects sizes? – Khaled Feb 18 '11 at 07:39 — Khaled, Feb 18 '11 at 07:39

score 17 · Accepted Answer · edited Nov 07 '15 at 00:04

Let's throw some numbers around:

1,000 at a time should be no problem at all, in general.
10,000 at a time will work but you have to know your entire setup intimately and never block any of the CPU's waiting for any I/O, as well as not be limited by your bandwidth. (assuming 10KB per http request and response @10,000 requests per second, that's 97 MB per second)

The rest of this answer is just some discussion to just point you in some direction

I've thought of this question a few times, and because you are looking for ball park figures then this might help you:

Looking at front end dedicated servers that handle a specific task (but talk to other services to do work for them, like the google search infrastructure which would call on many other machines to get pieces of information), then your front end machines that handle the actual client sockets should be able to handle about 15,000 users concurrently (ballpark, assuming binary protocol) because if you look at how many players play per server on WOW, depending on the strength of the machine, they have from 6-30,000 users per "world" - which I assume is one machine supported by other services at the back. I can't find the page now that lists servers and players per server. This isn't a web site though, but you are looking for ballpark information and this might lead you to the right place.

Have a look at wikipedia's server stats if you are interested in some architecture numbers - http://stats.wikimedia.org they show every detail you might be interested in regarding http requests.

Looking at IIS stats, take this site for example, stackoverflow.com - they were doing 16 million hits per month on 2 web machines and I'm sure more than 1000 people must have been on it at a time. There's an article on HighScalability.com about stackoverflow's servers. Update -- now at 95+ million per month.

SO has servers in other countries? – Motivated Student Feb 18 '11 at 09:27 — Motivated Student, Feb 18 '11 at 09:27

score 1 · Answer 2 · answered Feb 18 '11 at 07:49

1

How much wood would a woodchuck chuck if a woodchuck could chuck wood?

Your question is almost impossible to answer, even if you gave the complete server specifications along with the details of the data you are serving, such as if anything is dynamic, the average request size and such. The only way you can get a good answer is to setup the server, throw requests at it and measure what happens. That way you will also find your bottlenecks.

answered Feb 18 '11 at 07:49

pehrs

8,749
29
46

9

Is that an African or European woodchuck? – Evan Anderson Feb 18 '11 at 07:54
if the server has been setup, how to simulate the 1000 or more requests? – Motivated Student Feb 18 '11 at 08:43
1

Using load simulation tools, like Visual Studio has. Btw., a TX 100 X1 is not an average server, it is totally low end. This is like saying "an average car like a ford focus". It is also outdated- 8gb max ram? My workstations have that. – TomTom Feb 18 '11 at 08:48
@TomTom: OK. Let us assume the Ram 1 GB and one core 2.4 GHz, #rough concurrent connection? In hundreds or thousands ? – Motivated Student Feb 18 '11 at 08:57
4

Depends on what you do. Ths is btw not a server but a mobile phone. Get a real server. – TomTom Feb 18 '11 at 11:04

score -1 · Answer 3 · answered Jan 13 '14 at 19:01

Server (site process / DB ) can eat lot more than your (socket,TCP stack,Network adapter) unless badly designed. Better question is how menu simultaneous TCP connections can server handle.

On average linux system that critical point is 100 after degradation begins. This means it can handle more that 100 but not so efficient. It can push probably up to 180 after it begins to behave totally unacceptable.

Microsoft claims latest windows servers critical point is some ware around 500 by writing more efficient TCP stack.

But note this is a lot. If you have 150000 visitors per month that is around 5-6 simultaneous connections so if you don't have 10 times more visits you should worry about this especially if you host site on real server machine.

BUT, BUT, BUT BE AWARE!!!!!!!! sites are commonly hosted on virtual machines, and then web servers on each virtual machine can have multiple sites , so this accelerates way to critical point. Critical point I talk about stands for real machine - network adapter. In shared environment other users take their part of resources.

This is explanation why you get so poor experience when you host your site with cheep hosting provider. In order to cut expenses they tend to have as much as possible virtual machines on same real server with as much sites each can handle on other hand they have physical network adapter to push traffic for all of that.

where did You get this from ? "On average linux system that critical point is 100 after degradation begins." ~ 1995 ? completly outdated information or post some reference. — Bartłomiej Zarzecki, Jan 13 '14 at 19:26
its probably in some IIS book or reference, to make Microsoft look good. — code ninja, Dec 19 '14 at 18:31

How many simultaneous requests can be handled by a medium class server on the average?

Edit 1

3 Answers3