Archived from groups: microsoft.public.win2000.active_directory (
More info?)
If I remember the design characteristics correctly 300 users is the cutover
point. So if everything is about 300 or less than unless you have other
bandwidth issues you can centralize your design. You could throw some dc's
out there for high use and install dns and integrate it with AD, but your
plan will probably work ok. I would definitely have multiple DC's with
multiple DNS servers (Integrated is best) on separate subnets if possible.
--
Paul Bergson MCT, MCSE, MCSA, CNE, CNA, CCA
This posting is provided "AS IS" with no warranties, and confers no rights.
"Marlon Brown" <marlon_brown@hotmail.com> wrote in message
news:eeLNXPsOFHA.244@TK2MSFTNGP10.phx.gbl...
> Hello Paul,
>
> It is hub and spoke topology.
>
> 1 major site has 1,000 users.
> 10 major sites have between 150-300 users
> 9 minor sites have between 50-90 users.
>
> The problem is that our IT staff is very limited. My proposal is that we
> centralize the whole thing, take statistics of system downtime and
> re-address issues and place servers in the local branch offices if it
turns
> out telco downtime is more often then what we would like to.
>
> I wouldn't want to see any system downtime of course, but if it happens
once
> or twice a year for few hours, that would be acceptable in my environment.
> We are in the education business and people could survive in case of short
> downtimes.
>
>
>
>
>
>
> "Paul Bergson" <pbergson_nospam@allete.com> wrote in message
> news:e8ySWkqOFHA.3940@TK2MSFTNGP12.phx.gbl...
> > With this type of speed you can look at it as a single site and balance
> out
> > things like the whole system was all in one location.
> >
> > How many users, workstations, member servers exist in a branch office?
> How
> > important is a site? Can it be down from the rest of the system during
a
> > telco failure? Are there any alternate routes or is it a hub and spoke?
> >
> > There is more to this question than just the speed. But speed
definitely
> > makes things much easier and simpler.
> >
> > --
> >
> > Paul Bergson MCT, MCSE, MCSA, CNE, CNA, CCA
> >
> > This posting is provided "AS IS" with no warranties, and confers no
> rights.
> >
> >
> >
> > "Marlon Brown" <marlon_brown@hotmail.com> wrote in message
> > news:#j7ChQiOFHA.624@TK2MSFTNGP10.phx.gbl...
> > > We have an option to replace current T1 circuits by fiber for all 30
> > branch
> > > offices.
> > >
> > > Branch offices would be connected like this:
> > >
> > >
> > >
SmallBranch1-------fiber------>MediumBranch---------fiber---->DATACENTER
> > > SmallBranch2-------fiber-------|
> > > SmallBranch3-------fiber-------|
> > >
> > >
> > > In the slowest link, the fiber link should delivered 100MB/s. Fiber
> should
> > > be significantly more reliable than T1's. Of course we still have
> chances
> > to
> > > see a fiber cable getting disrupted or something like that.
> > >
> > > Would you put DC/DNS/DHCP servers on the remote offices, or generally
> for
> > > fiber you centralize the whole thing on the main data center ?
> > >
> > >
> > >
> > >
> >
> >
>
>
Facebook
Twitter
Instagram