|
Home > Archive > IIS Server > June 2005 > IIS 6.0 Scalability
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
IIS 6.0 Scalability
|
|
| Aragorn 2005-06-28, 5:54 pm |
| Has anyone done tests and or have whitepares describing the preferred
scalability model for IIS 6.0? Scaling out vs. Scaling up....
--
____
- Aragorn
| |
| David Wang [Msft] 2005-06-29, 2:48 am |
| I don't have the public papers handy, but we definitely did significant work
on tuning IIS6 to scale on 1P, 2P, 4P, and 8P. IIS5 pales in comparison.
Things like state-management, contention, etc are more likely problems, but
those are not necessarily issues with IIS 6.0 scalability -- they are
usually issues with the application code that need to be dealt with in order
to properly scale.
So, the real question is whether your code scales out or up. :-) If you find
IIS 6.0 to be the bottleneck, we'd like to know about it.
--
//David
IIS
http://blogs.msdn.com/David.Wang
This posting is provided "AS IS" with no warranties, and confers no rights.
//
"Aragorn" <Aragorn@discussions.microsoft.com> wrote in message
news:62664CC9-099F-4D85-A43B-5112AB986119@microsoft.com...
Has anyone done tests and or have whitepares describing the preferred
scalability model for IIS 6.0? Scaling out vs. Scaling up....
--
____
- Aragorn
| |
| Aragorn 2005-06-29, 5:53 pm |
| Hi David,
I'm not questioning the ability of IIS to scale. What I'm looking for is
facts and figures regarding the factor of scalability. I'm looking for the
"You will get 50%-80% more throughput on IIS if you scale from 1 to 2
processors but I can get up to 200% more throughput if you scale from 1 -
1cpu to 2 1cpu servers in NLB" for example.
What is the "Best Practice" or even best "Proven Practice" when it comes to
scaling IIS?
--
____
- Aragorn
"David Wang [Msft]" wrote:
> I don't have the public papers handy, but we definitely did significant work
> on tuning IIS6 to scale on 1P, 2P, 4P, and 8P. IIS5 pales in comparison.
>
> Things like state-management, contention, etc are more likely problems, but
> those are not necessarily issues with IIS 6.0 scalability -- they are
> usually issues with the application code that need to be dealt with in order
> to properly scale.
>
> So, the real question is whether your code scales out or up. :-) If you find
> IIS 6.0 to be the bottleneck, we'd like to know about it.
>
> --
> //David
> IIS
> http://blogs.msdn.com/David.Wang
> This posting is provided "AS IS" with no warranties, and confers no rights.
> //
> "Aragorn" <Aragorn@discussions.microsoft.com> wrote in message
> news:62664CC9-099F-4D85-A43B-5112AB986119@microsoft.com...
> Has anyone done tests and or have whitepares describing the preferred
> scalability model for IIS 6.0? Scaling out vs. Scaling up....
> --
> ____
> - Aragorn
>
>
>
>
|
|
|
|
|