01-31-05 10:53 PM
Hi Chris,
I'm also an Escalation Engineer.
Would be nice if you could send me the service request number you had and
the name of the EE to my email address: webmaster@stefan-gossner.de
Cheers,
Stefan.
"ccuthbert" <nsp_skier@hotmail.com> wrote in message
news:1107182343.286759.239010@z14g2000cwz.googlegroups.com...
> Stefan,
>
> Interesting reply. I spoke with a Microsoft escalation engineer and
> that is not what I was told. I was specifically warned against using
> host header mapping, and I was also given the limit of 15 channels
> under the root and 300 objects in any container regardless of whether
> there were only MCMS sites, which is the case in my system. All sites
> in my system on the server are MCMS managed sites. The MCMS
> documentation: Performing Administrative Tasks -> MCMS Performance
> Considerations -> Best Practices for MCMS Site Performance states:
>
> Site Manager best practices
> Limit the depth of container hierarchy.
> Limit the number of containers under the root node. (For example,
> channels should not have more than 10 to 15 immediate children
> containers.)
> Limit the items in a container (to about 300).
> Distribute items over multiple containers, to ensure that the number of
> items in each container does not exceed 300.
>
> While I certainly would defer to you, you've have helped me with many
> problems I have had in the past, I did find that performance was
> measureably better after I followed the advice of the escalation
> engineers. I agree 100% you should always do load testing as this is
> the only method to ensure you have an accurate idea of the load
> capability of your site(s).
>
> The method I am using was reccomended to me by the microsoft engineers
> after multiple conference calls over a 1 week period.
> As always thanks for you reply
> Chris
>
[ Post a follow-up to this message ]
|