Hello gang,

I am in the process of setting up a new Windows 2003 AD environment and one
of my goals is to provide as much redundancy as possible right from the start
as we are a very fast growing company and this environment is being built
with scalability and growth in mind.

My question is this. As we migrate old file servers and consolidate the
data on a SAN I would like the option to have redundancy from a file access
perspective. What I am wondering is whether implementing a distributed file
system will allow me to have two or three primary file servers to act as the
front end for users while the data is stored in SAN pools. I want the option
to be able to take down any one of the file servers at any time while
allowing users continued access to the data stored on the SAN. This will
include the option for patch updates during working hours, or in the case of
hardware failure. The SAN storage is split across mulitple enclosures
connected to mulitple switches so I am trying to emlimiate any one point of
failure.

I am also aware that clustering will service this purpose as well. My
question around clustering however is for the cost associated with purchasing
Enterprise Server (if decide to utilize 2003 which is quite likely,
admittengly I am not sure of the price difference between between 2003
standard and 2003 enterprise) and the configuration and administration of a
cluster vs. using DFS (if its feasible) what route would be the best approach.

Advice? Opinions?

Thanks kindly.

So, what are SAN pools?
Is this a block level data mirroring solution?
If so, then you don't need clustering.
A DFS deployment with multiple link targets will do.



--
Glenn LeCheminant
CCNA, MCSE 2000/2003 + Security

"a_user" <(E-Mail Removed)> wrote in message
news:AF6E7FF9-7AD0-4977-9F96-(E-Mail Removed)...
> Hello gang,
>
> I am in the process of setting up a new Windows 2003 AD environment and
> one
> of my goals is to provide as much redundancy as possible right from the
> start
> as we are a very fast growing company and this environment is being built
> with scalability and growth in mind.
>
> My question is this. As we migrate old file servers and consolidate the
> data on a SAN I would like the option to have redundancy from a file
> access
> perspective. What I am wondering is whether implementing a distributed
> file
> system will allow me to have two or three primary file servers to act as
> the
> front end for users while the data is stored in SAN pools. I want the
> option
> to be able to take down any one of the file servers at any time while
> allowing users continued access to the data stored on the SAN. This will
> include the option for patch updates during working hours, or in the case
> of
> hardware failure. The SAN storage is split across mulitple enclosures
> connected to mulitple switches so I am trying to emlimiate any one point
> of
> failure.
>
> I am also aware that clustering will service this purpose as well. My
> question around clustering however is for the cost associated with
> purchasing
> Enterprise Server (if decide to utilize 2003 which is quite likely,
> admittengly I am not sure of the price difference between between 2003
> standard and 2003 enterprise) and the configuration and administration of
> a
> cluster vs. using DFS (if its feasible) what route would be the best
> approach.
>
> Advice? Opinions?
>
> Thanks kindly.