camargo@cs.columbia.edu (Kiko) (05/27/91)
Hi there,
Well, here we go again...
We are ready to start deploying Thin-Ethernet cables in a 3-floor building,
and I just came up with a few concerns that might not be relevant. In any
case, here they are:
. What is the minimal "T" cable drop, from a Thin-Ethernet to the NIC ?
. Is is really advantageous to have the T-connectors in the NIC's
themselves ?
. What is the effect of using wall-plates with 2 BNC connectors, and
each workstation connected to the ethernet with a T in the NIC ?
Won't this arrangement easily exceed the "minimal number of
connections" ?
We are planning to run 6 sub-segments, each about 120 feet long, with 10
workstations in each (in average). They will be connected through a T-connector
in the back of 3 NIC's in the file-server, thus yielding 3 actual segments.
The realm of the question is: Should we put "T-connectors" inside the wall, or
have the "T-connectors" in the workstations? While the "T-in-the-wall" solution
is "cleanner", the "T-in-NIC" is more flexible, but adds to the number of
connections in each segment.
My knowledge of "Thin-Ethernet limit-values" calls for 30 connections, at most,
in each segment, limit that will be easily exceed if we use the "T-in-NIC"
solution. Is anybody out there to say that I'm wrong ?
In case I'm not, how far away can a workstation be from a "T-in-the-wall" ?
Well, thanks for any help.
/Kiko
camargcs.o@colbia.edu
[Disc: To this date, I had experiences only with the "T-in-NIC" solution, in
much more limited spaces. Thus, these questions were never an issue.]hamish@waikato.ac.nz (05/27/91)
In article <CMM.0.88.675297173.camargo@cs.columbia.edu>, camargo@cs.columbia.edu (Kiko) writes: > Hi there, > > > Well, here we go again... > > We are ready to start deploying Thin-Ethernet cables in a 3-floor building, > and I just came up with a few concerns that might not be relevant. In any > case, here they are: > > . What is the minimal "T" cable drop, from a Thin-Ethernet to the NIC ? > > . Is is really advantageous to have the T-connectors in the NIC's > themselves ? > Ill take these 2 together. In thin-net it is mandatory to have the T connected directly to the NIC. Any length of cable will produce echoes and may well render the network unusable. having said that I have on occasion in an emergency used a short length of cable straight to the NIC, but it didn't work very well. So stoick to the specs and put the T directly onto the NIC. > . What is the effect of using wall-plates with 2 BNC connectors, and > each workstation connected to the ethernet with a T in the NIC ? > Won't this arrangement easily exceed the "minimal number of > connections" ? > Ahh? If I understand the question properly, then make sure you don't go over spec on the number of connections allowed. Each wall plate will act like a T is there (pretty much anyway) whether there is a T or a straight piece of cable. I have had experience with AMP connectors (something DEC seem to be getting into) that you simply push a special T cable into that breaks the segment and routes it through the double coax to give a T on a single cable. Nice and neat but can be unreliable if connected and disconnected roughly too many times. (ie the wall connectors are pretty fragile) > We are planning to run 6 sub-segments, each about 120 feet long, with 10 > workstations in each (in average). They will be connected through a T-connector > in the back of 3 NIC's in the file-server, thus yielding 3 actual segments. > > The realm of the question is: Should we put "T-connectors" inside the wall, or > have the "T-connectors" in the workstations? While the "T-in-the-wall" solution > is "cleanner", the "T-in-NIC" is more flexible, but adds to the number of > connections in each segment. > Put them ON the NIC. In the wall wont work with thin net. > My knowledge of "Thin-Ethernet limit-values" calls for 30 connections, at most, > in each segment, limit that will be easily exceed if we use the "T-in-NIC" > solution. Is anybody out there to say that I'm wrong ? > > In case I'm not, how far away can a workstation be from a "T-in-the-wall" ? > 0. See above. Hope this helps. -- ============================================================================== | Hamish Marson <h.marson@waikato.ac.nz> | | Computer Support Person, Computer Science Department | | University of Waikato | |Disclaimer: Anything said in this message is the personal opinion of the | | finger hitting the keyboard & doesn't represent my employers | | opinion in any way. (ie we probably don't agree) | ==============================================================================