Note: if this would be better placed in a different section of this forum for best responses, please advise. Ty
I need to figure out how to design and number a mixed 100/1000 MB client environment.
The mix:
• 28 G3 PPC clients only have 100 MB capable nics.
• 15 G4 PPC clients or so have 1000mb nics
Resources:
• Server G5 tower, 8gb ecc ram, Highpoint RocketRaid 2320 card
• enclosure with the share points in Raid 5
• Two independent nic ports
• One wireless nic
Current design:
• Eth 1 feeds all the g3 100 mb capable clients all of which have fixed ip/s
• The server is the router numbered 192.168.1.1 255.255.0.0
• Clients are 192.168.1.2 255.255.0.0 and so on.
• All clients are wired through cascaded (stupid) switches that terminate with
machine 27.
• Currently there is no common switch between the 1000 cascade and the 100
cascade.
Needs:
• All clients 100 and 1000 will need access to the same shares I host.
• All clients must have fixed IP/s
• I would like to be able to save my EN 0 for my own high bandwidth Internet
connection.
Main question flow:
• If I plug in the eth 1 feed into the first of my 1000gb switches (which are new
auto negotiate types) instead of directly into the eth 1 port in the tower:
1. Will that create a loop back?
2. If not, will it slow down my 1000 mb branch
3. If not, should I just pick up numbering the 1000 MB clients 192.168.1.28
and so on. 😀
Comments are closed